Changelog

Firmware

1. docs: a stock defect worded so site search still puts Help's answer first for a mouse that won't move07d5c95
2. ci: the firmware zip goes on the GitHub release as well as the Discord post87db760
3. docs: release notes for 3.4.6d6171e2
4. tools: gate docs, rig notes and test comments read on first pass24db7a5
5. firmware: comments read on first passc349253
6. docs: the spec, write-ups and comparison read on first pass, with symbols for counts and limitsd89e5ac
7. tools: gate docstrings, rig notes and test comments in the project's words77de5dd
8. firmware: comments in the project's words, and the ones that had the code wrong86e8c59
9. docs: the protocol spec, architecture and comparison write-ups in the project's words, and the facts they had wrong3ada641
10. docs: the suppression gate needs a device that sends nothing unprompteded20cb8
11. docs: the protocol doc, README, PSRAM page and detection notes follow 3.4.6cdfd496
12. docs: comparison rerun on the final build, with report latency, the lead and bulk data on Medius brought up to dateb6695e9
13. docs: the design record, release notes and gate list follow the bulk data fix9daf47e
14. firmware: a device's bulk data reaches the PC whole, however late or slowly the PC reads it7457c41
15. docs: the Discord foundation's rules post ends on rule 53b5f48e
16. docs: the Discord rework's embeds and panels in the site's blue, and the comparison's benchmarks9193c6d
17. tools: the release post's embed is in the site's blue, its title linking the release97475a5
18. docs: the Discord rework's embeds follow the bot's lookbc4e32a
19. tools: the release post's text is its pings, the notes an embed titled with the release967f32b
20. docs: the Discord rework's bot posts are embeds503f9f0
21. tools: the release's Discord post is an embed, with the ping and the credits' thanks in its text2b78491
22. docs: the design record, release notes and bench notes follow the 400 µs lead and the shorter wait after a late reporta302c08
23. firmware: every report the box relays is readied ahead of its other work, not only the mouse's173dc14
24. docs: the Discord rework adds a comparison channel holding the site's summary3bde844
25. firmware: one late report no longer leaves the reports after it a poll late for secondsfdcab3f
26. firmware: the mouse's reports keep their poll under heavy load, and reach the PC 50 µs sooner5044c5d
27. docs: comparison has no pending tests left; the three that need Windows are listed as not covered47ca3c9
28. docs: comparison counts reports carrying only injected motion beside a varying mouse, and shows each clone's endpoint 0 reply timedf7ed4e
29. docs: comparison lists stock's 1 ms mouse interval as a defect on every row it wins02a91e7
30. tools: comparison checks take a defect on every row it decides, and the device matrix has 2 ms and 4 ms mice263e809
31. docs: comparison report latency rerun, p99 in place of the slowest reportf3746bc
32. tools: report latency script places the emulator's clock per window and follows its held-back intervals5953efb
33. docs: comparison report latency on both boxes at 125 Hz to 1 kHz4163514
34. tools: comparison script timing every native report from the emulator to the PC, on both boxes401aab8
35. docs: comparison click timing rerun on the clip fix0b2d552
36. firmware: a clip's clicks and keys reach the PC as far apart as they were played, the first after a pause included328db02
37. docs: comparison rebuilt on one hub with Medius 3.4.6 final, the pending tests measured on both boxes7751eb7
38. tools: comparison scripts give Medius the same fast and back-to-back legs as stock, and keep a clone that reconnectseb32686
39. firmware: every interface number, setting and endpoint a device has is driven, and a stream polled every frame keeps its rate beside a slower one43f9538
40. tools: comparison scripts for interface data, intake limits, typing, sleep and timing legs; same-hub evidence and its manifest8ddbc03
41. docs: v3.4.6 sleep, halt, render debt and link fixes in the design record, protocol and release notes2f353a4
42. tools: gates find this box's clone by its USB port as well as its ID, and the suspend and halt gates check more1c7907a
43. firmware: the PC's sleep and a device's halts pass through, rendered debts are delivered, and a device change loses no link frameaa833ee
44. docs: v3.4.6 faithfulness fixes design record and release notes0d37dfd
45. tools: a clone lookup never takes another box's clone25c7c2b
46. firmware: an endpoint the mouse stops reaches the PC, and the device chip keeps one snapshot825a402
47. firmware: rendered moves deliver exactly what was sent, with no count against them49edc1d
48. docs: comparison results rebuilt as direct like-for-like rows, with defects and pending testsd03373e
49. docs: comparison results: relay reruns, Equal by overlapping ranges, defects that reproduce, pending tests for readers8b70cef
50. docs: comparison smoothing group described in fewer wordsb288474
51. docs: comparison results format: unmeasured chart points, defects on rows of results, stricter checks; seed refreshedda7b336
52. docs: comparison results in one checked file, for the site1b87cad
53. docs: comparison section designdbdc893
54. docs: dashboard controls design recordf7f2945
55. docs: v3.4.6 render totals and query replies36d86fc
56. tools: wire-level timing for the comparison, and the transform gate reads a reader overflow as blocked8cb9f99
57. firmware: the control port answers queries soonera381d09
58. firmware: rendered motion keeps its totals with every report the model's375710d
59. release v3.4.60e0175b
60. docs: v3.4.6 design record, release tables and protocol change log8ac0726
61. tools: the transform gate reads its own box's cloneed59ffc
62. firmware: devices enumerate on the USB specification's reset timingsb93a834
63. firmware: typed strings arrive whole, and bursts of commands are keptde854a4
64. firmware: rendered moves add up while the mouse moves2cd6524
65. docs: comparison wording4229ddc
66. docs: comparison refusals with the mouse moving, readers in their own processes, hub timingc65622c
67. docs: comparison move sizes, text clumps and the reworked lock check on stockb728b24
68. docs: comparison move acceptance by form and pacing, lock checks, native rate triggersc055649
69. docs: comparison stock re-runs after power cycles, native rate while injecting, parser trigger6875489
70. docs: comparison second review: trigger order, interval rule, render evidence, imperfect coveragef0f1ff8
71. docs: comparison data path checksfad19f0
72. docs: comparison report descriptor limits and endpoint 0 shapes4a6c80c
73. docs: comparison corrections, Medius fastest-settings and cold-start runs, stock re-run scriptea6672b
74. docs: the stock firmware comparison: rig, measurements and researchbff4704
75. tools: quiet_clone and clone_guard cover two clones on one PC9ee97ab
76. docs: the dashboard's Flash page and one-count flash meterb6d4856
77. docs: the ROM client spec's erase before the write and its smaller blocks970fa9c
78. docs: the device feed's first post and the changelog's history in the Discord designe2c8c4e
79. docs: the ROM client spec's stock done screen, one lineba5c3ae
80. release notes: a refused Discord post prints Discord's answer0c4a7a2
81. docs: the dashboard's boot ROM client, spec and planb4a289d
82. docs: release posts by Clanker and the device feed in the Discord design79bd102
83. release notes: the Discord post comes from the server's bot, not a webhook60fb701
84. docs: the honeypot and bot settings in the Discord spec8562eb9
85. docs: Home under Support in the Discord layoute87073b
86. docs: the Server Tag in the Discord checklist36866d9
87. docs: what the Discord spec holds now08f9cad
88. docs: the Discord Apply checklist3507bba
89. docs: options spec, the stock erase in regionsffb0cfd
90. release notes: ping a release role in place of @everyone when one is set414f00b
91. docs: Discord onboarding beside the captchad29730d
92. docs: place addresses spec and plan, as built91c0a9c
93. docs: Discord layout after the second reviewddca522
94. docs: Discord rework after review5a7d5c7
95. docs: options spec, project termsfc385e5
96. docs: Discord rework design0623ddf
97. docs: options spec, details as tables31b5023
98. docs: rule 2 is the rule as written2299430
99. docs: options spec, every option a section6a04047
100. docs: options spec, links both ways and plain sectionsec9691d
101. docs: Discord foundation spec as builtb1d3e86
102. docs: options page and dashboard fixes spec and plan, as built5b95e15
103. docs: Discord foundation spec, build changes075af56
104. docs: Discord foundation plan; spec binds by name1b9bfe9
105. docs: Discord foundation spec7296017
106. docs: deferred minors spec and plan, as built0de35d1
107. docs: copy link spec as built1ae6b55
108. docs: copy link spec and plan6431caa
109. release notes: no preview for any other link in the postf2e8b1a
110. release notes: link the release's page, so its card showsd349667
111. docs: plan wordingef24270
112. docs: link preview plan; theme-color set before first paint95bfe26
113. docs: link preview spec and the approved cards975886b
114. docs: site loading spec as built; search spec follows it5127cd9
115. docs: site search spec as built049759f
116. docs: site search spec and planbe54efa
117. docs: site surfaces implementation planac49efd
118. docs: surfaces mockup, approved version5a77bb8
119. docs: surfaces mockup, eighth version4f21c4e
120. docs: surfaces mockup, seventh versionda90739
121. docs: surfaces mockup uses the full page width0910fcc
122. docs: surfaces mockup, fifth versioncad1657
123. docs: surfaces mockup, fourth version7a3f54c
124. docs: surfaces mockup, third version21fe54e
125. docs: design mockup for the site's remaining surfaces9951773
126. tools: closer injection bursts in the site feed capture, and no capture dateeb72838
127. tools: site capture rigs read the run exactly and restore every option they changefc314f6
128. tools: descriptor and report capture for the docs sited01f12c
129. docs: plan wording0a95b59
130. docs: site theme and user pages plan8c13329
131. docs: site theme and user pages specc4a6fe7
132. docs: the approved site design mockupc213605
133. docs: every dashboard page is indexed8edef13
134. docs: the search foundation spec as built35d4814
135. tools: release notes refuse unknown sections and stray mentions; long posts split cleanly42821b3
136. docs: wording in the search foundation spec and plan371a670
137. docs: release notes for v2.2.1 to v3.4.5d8ea65d
138. tools, ci: release notes written once, posted to GitHub and Discorda572a8e
139. docs: plan for the docs site's search foundation304274c
140. docs: design for the docs site's search foundation6b7013f

Library

1. docs: doc comments, the C header and Python docs read on first passc8dc355
2. docs: doc comments, the C header and the Python docstrings in the project's words, and the facts they had wrong88f2073
3. library 3.4.6: close a stream from another thread, the C ABI on crates.io, and docs that match the code5260727
4. docs: the library follows firmware 3.4.6: what relay_drops counts, sleep, STALLs, typing capacity and the RAW IN queue042a40c
5. library: mock refuses an activate or reboots after one, tests for the abort and the confirm retry390deeb
6. python: Device and ClipHandle close safe across threads13e2dd3
7. python, capi: closer to the crate, safe stream close across threads, firmware update and confirm through the crate881a342
8. capi: MediusDevice::into_raw, a handle over a Device that's already open0d7e773

Docs

1. comparison: data synced; search for a mouse that won't move finds Help's answer again017498a
2. comparison: a count out of a total reads 6/10a109fb7
3. site: every page reads on first pass, with symbols for counts and limits30cfe39
4. docs: comparison wording synced from the firmware repo: native rate, emitted reports, relaying6335da1
5. docs: every page and the dashboard in the project's words, and the facts they had wrongd72fe58
6. docs: library 3.4.6: closing a stream from another thread, firmware update and confirm in every language, and the mock's update knobscf7e706
7. docs: the site follows firmware 3.4.6: sleep, STALLs, typing capacity, what relay_drops counts and when a transfer gets no answerec1cbbc
8. test: the stats page's dates are checked in the locale the tests run under2df307e
9. docs: the comparison follows the final firmware build, and the RAW page describes how bulk IN fills its queue066ed8b
10. feat: the comparison JSON carries a few benchmarks with both boxes' values, and every area's pagef8a3694
11. feat: the comparison JSON counts the results each box is ahead in, by area487b38a
12. fix: the comparison JSON is built once per deploydba8df8
13. feat: the comparison's Overview as JSON, for the Discord botae4be1a
14. comparison: the Method page drops its pending tests sectione4f07d6
15. comparison: injected-only reports beside a varying mouse, and endpoint 0 reply timescdbcf84
16. site: links land clear of the bar and light the place they go to, and the settings table's headers wrap over their columns0c03139
17. comparison: a defect marks every row it decides and links to its own place in the listb521746
18. comparison: report latency rerunac08092
19. comparison: report latency section, led on the Overview6c7aca3
20. comparison: click timing on the clip fix86c7316
21. comparison: results from the one-hub final pass, the pending tests added, and the rig described as one hub9fb2033
22. docs: comparison pages rebuilt as direct like-for-like results; headings and in-page links fixed682bbeb
23. docs: comparison results refreshed; Method states the Equal rule and what interval sd measures4412cf5
24. feat: a Comparison section: MAKCU's stock firmware against Medius, area by area7c3acdc
25. docs: the privacy page carries what the law asks for, in the site's voice5169636
26. fix: rate limits drop an address when its window ends, and count each visitor rather than the proxyfbba2e2
27. feat: a Flash page that writes either chip through its ROM, older releases, and dashboard controls that match the sited539a57
28. docs: update firmware version strings to 3.4.65a216b0
29. test: comment wording in the Devices JSON test405f9d9
30. test: each device's address in the Devices JSON is its own row's carda0aabd5
31. fix: the privacy page dates a scam signature from the last scam post that added it30c425e
32. feat: each device in the Devices JSON carries its card's addressc1dca68
33. feat: the Devices list as JSON, for the Discord bot21eb8fa
34. fix: the privacy page keeps scam signatures for a year, as the bot does161c3f2
35. fix: the privacy page names scam signatures as the bot keeps them50b709d
36. feat: a copy of a box's stats on request; the privacy page covers copies, verifications, signatures and DMed transcripts875c68e
37. fix: the stock firmware relay follows MAKCU's toolkit 1.15.20b9fe229
38. fix: the stock source reads as the release does, plainer fix offers, a stock erase as MAKCU's toolkit does it, and a fuller privacy page4fdc7ba
39. feat: privacy page, a footer link to it, and a stats delete commanded419c8
40. feat: an address and a link card for every section, tab and panelc210347
41. docs: emit, ride and relay in place of paraphrases across the sitee28100b
42. docs: option details as tables with measured delays, not proseda96a00
43. docs: Rendered motion and Wire rate as sections of the Options guide2ebbb90
44. docs: plain option explanations with links back to the dashboard, and noun headings across the reference13ba603
45. docs: plainer wording across the reference pages, comments and test names24c9f54
46. fix: review findings for the Options page, the fix offers and the stock firmware relay and flashd65f947
47. docs: Help sends owners to the dashboard for older boxes and MAKCU's stock firmware63903b1
48. feat: Update's Manual tab can put MAKCU's stock firmware back on either chipa08c74d
49. feat: the server relays MAKCU's signed stock firmware, checked as MAKCU's toolkit checks itd3287ee
50. feat: the dashboard offers the fix a connected device needs, in placebef0948
51. feat: boxes on firmware from v3.2.0 get the options their firmware hasb83f9fe
52. feat: an Options page in the Guide, linked from each option on the dashboard864ed80
53. fix: review findings: a poll that stalls after a flash, the Devices twin's links, tab spacing per strip, and the feed on browsers that round97be654
54. fix: no page scrolls sideways at 320 px, tab strips fit before they scroll, and Copied stays clear of tabs8c92a13
55. fix: the landing's feed and descriptor are right before the app runs, pause under the pointer, and take the keyboard2aa8ac3
56. fix: a link to an option says how to reach it before a box is connected0de7f3c
57. test: crawlcheck fails a doubled or missing link-preview tag, and one naming another pagec7b11a7
58. fix: link cards draw every Latin letter off the main thread, item pages carry their own data, and drafts stay off the changelog33762e7
59. fix: an unplugged box waits for its port without timers, and Reconnecting reads as a statusaf7a5ea
60. fix: link icons take no room until shown, keep headings' names, and Copied never widens the pageb921cb9
61. docs: the link icon and Anchor in CLAUDE.mde068230
62. fix: the open tab's Copied stays inside the tab bard50cf64
63. test: the build fails on a place search can land on with no link icona2880ea
64. feat: copy the link to the open dashboard tabfc844d6
65. feat: copy the link to a Help answer, device or release4d5c7ba
66. feat: a copy link icon on every section, panel and marked blockaa606cd
67. fix: one release date for every reader, page and cardd884d2a
68. docs: Pulsar ZywOo The Chosen Mouse reported working675f071
69. docs: six more devices reported working, four confirmed on v3.4.585a9319
70. fix: link cards kept by what they show, unknown items stay a 404, release lookups answer within 5 s513e5fb
71. test: crawl every link card and item address0cfbc07
72. fix: fill the changelog and stats blocks as the prerender now captures them683aef8
73. feat: serve link cards and item addressesc596cf4
74. feat: link card tags on every page, and addresses for one Help answer, release or device6ab40f3
75. feat: draw link cards as PNGs on the serverbc87a87
76. feat: what each link card says, and addresses for single items71ff1ec
77. feat: link card fonts and text layout798032a
78. fix: one top bar and grid for the site, scroll kept per page, motion that never plays twiceda5197e
79. perf: split the site's code per page and fetch search on use2653d39
80. feat: site search built from the pages themselves514fb1c
81. feat: rebuilt site search1b90eb8
82. fix: the box diagram matches the boarda298aa7
83. fix: the safety clear's caption is one short line1845204
84. feat: one motion for every page0fb5d52
85. fix: the dashboard's rare states, each rendered and measured07ef510
86. fix: the 404's description names the links it carriesdf1517d
87. fix: the 404 page prerenders again237b076
88. fix: dashboard controls as designed: sliders, steppers, buttons and linksaace53a
89. fix: review findings across the dashboard and the Guide432b849
90. docs: pages open on their title and first paragraph, with no subtitle465420c
91. feat: the 404 names the address asked for and where to go instead55e44e2
92. feat: the Guide in three pages: Start here, Devices and Helpe2ceb4c
93. feat: Changelog with each version beside its notes, and commits grouped by repo418a637
94. feat: Stats on tabs, with charts that read out the column or row under the pointerbade974
95. feat: Control on tabs, Update with a manual flash tab, and Set up as one flowc05ee44
96. feat: the Device page on tabs: overview, options and a log to copy or save8211222
97. feat: the box drawn from its board, for wiring and install steps07b0009
98. feat: segmented choices, a filterable dropdown and a log for the dashboarddd94a12
99. feat: page tabs and panels for the dashboard, and pages that use the full widthf035a3b
100. chore: a box that answers without hardware on the dev server25bcb11
101. feat: the landing feed shows a burst of injected motion every few seconds, and no capture date6a26bc8
102. feat: the landing feed recaptured with exact injected-byte marks9278d64
103. fix: the landing intro carries on when the app takes over, and descriptor panel fixesa6f6ce7
104. fix: landing and changelog answer within seconds while GitHub or Discord hang0130bec
105. fix: plainer Troubleshooting lead and search descriptiona3fe643
106. fix: the custom scrollbar shows in Chrome, square and blue on hoverba7b2c6
107. fix: the landing feed keeps every byte of a long report on screen at any widthbfec47c
108. test: crawlcheck holds every page to one h1 and a footer, and no hidden reveals1fa1d71
109. feat: landing page3967f7d
110. feat: captured descriptor and report samples for the landing page4e62486
111. fix: phone panels, scrolling, prerendered reveals, contrast and search anchors85f94ad
112. fix: phone page panel closes on a box pick or search, and dashboard tools get the full width9765e4a
113. fix: guide corrections, and page leads without trailing periodsc1c3e36
114. fix: compatibility rows stay apart per device, and the Discord count retries after a failure94c9e64
115. fix: prerender waits for the 404 page's header3568738
116. docs: project notes for the new shellf4613ce
117. feat: not-found page on the new shelld049c04
118. feat: landing page blocks1e3d911
119. feat: live figures on the landing and compatibility pagesd54320d
120. feat: guide pages for box ownersc6a3a05
121. feat: compatibility data1d3f078
122. feat: FAQ data and structured data95b7570
123. feat: docs content in the site themedf74a74
124. feat: frame diagrams as byte strips19b6a89
125. refactor: docs pages use page headers and sections5f893f5
126. feat: dashboard in the site theme6f16d46
127. feat: docs pages on the new shell20b6382
128. feat: page header, sections and on-this-page raildf5300b
129. feat: docs sidebar1bdd0e5
130. feat: site nav and footer4c7d086
131. feat: guide section routes99dc575
132. feat: scroll reveal helpersf5e97d9
133. feat: site theme tokens, fonts and base stylesf59b667
134. feat: the Control and Advanced control pages are indexedb1a7aa7
135. fix: the server's shared modules ship in the image; live pages answer 503, not a placeholder3050201
136. chore: wording in comments, a description and a test name0a59366
137. feat: bold and code in release notes; crawl check typechecks4905b9e
138. docs: the route registry, sidebar links and 404 in CLAUDE.mdd490c73
139. ci: sitemap dates from each page's last commit0df48e2
140. feat: release notes on the changelog, commits behind Show commits, filled in for crawlers02a5489
141. test: drop the removed unsupported-browser cards from the search exemptions9e34f4e
142. feat: dashboard pages keep their content in a browser without Web Serialb8349a0
143. feat: breadcrumbs and a site footer452aeae
144. feat: a sidebar of real links9598337
145. feat: a 404 page67d43ae
146. feat: 301 to the canonical URL and a real 404 for unknown pathsd1e876d
147. feat: page head and structured data from the registryff7fb51
148. feat: prerender every route, the dashboard included, from the registry6ab9ba9
149. feat: one route registry for page titles, descriptions and navigation13c9fe7

Firmware

1. docs, tools: v3.4.5's control request timing and RAM figures6ad9805
2. firmware: control requests are answered from RAMbf6e62f
3. docs: v3.4.5 in the release table and protocol change log2dcc498
4. release v3.4.595a4d53
5. docs: internal RAM on a box without PSRAM, and how the update gate checks it977e03e
6. tools: the update gate runs the device image as a box without PSRAM does2339868
7. firmware: the clone starts on a box without PSRAMe8c1e2f
8. docs: the usage stats design and plan reach the page from the Stats tab alonec2a431d

Docs

1. docs: update firmware version strings to 3.4.5116031c
2. dashboard: fewer words on the Update, Advanced and Device pages8b0a097
3. dashboard: the connect card without the stats line45c272b

Firmware

1. docs: the usage stats design's result names and totals4020244
2. docs: the usage stats page down to its numbers5fb9415
3. docs: the usage stats page is reached from the dashboard alone14867a0
4. docs: the usage stats design's limits, active boxes and shared devices7c2b211
5. docs: the public usage stats design and plan628d511
6. docs, tools: plainer wording in the protocol doc, designs and gate notes3dc6e29
7. tools: leave the control port readable by the browser dashboard390ca7d
8. firmware: the control port drops a frame its sender abandoned83fbd1b
9. docs: the USB2 flash design's rare paths and what they changed4a65e0d
10. docs, tests: the stable update path names its resync, waits and the messages between the chips8183489
11. docs: the USB2 flash design as built, and its hardware run19410d8
12. docs: name the stable update path section, and plain punctuation in the plan8df2fee
13. docs, tests: the update path keeps its shape from protocol 5 on9d7e587
14. docs: plan for flashing over USB2 from the dashboard's Advanced pagedd5c6ce
15. docs: design for flashing over USB2 from the dashboard's Advanced paged543602
16. docs: USB device box icon, and dropdown menus idle once closedbe31119
17. docs: box icons in the multi-box dashboard design6dd0725
18. docs: multi-box dashboard design, Add box tab and card wordingaa5f821
19. docs: the multi-box dashboard design as builtdfe9a8c
20. docs: design and plan for managing several boxes from the dashboarda7b7de2
21. firmware, tests, tools, docs: an imperfect clone keeps every IN endpoint lived4eb9ec
22. firmware, tools, docs: a lock is in force before any command sent after itea2e07a
23. docs: v3.4.4 in the protocol change log4d8f564
24. tools: the Logitech fix runs on Linux and Windows with nothing to installc6d64e1
25. docs: v3.4.4 in the release tablesd2f2839
26. firmware, tools, docs: an endpoint polled every frame keeps one place in it beside slower onesff60509
27. firmware, tests, tools, docs: the native rate reads true through reports a poll late74558cf
28. release v3.4.4c2417ef

Library

1. library: over_capacity says what it counts, now that the box serves every IN endpoint610a5fe

Docs

1. dashboard: Waiting for injection and Move the mouse to start as neutral chips292d018
2. dashboard: stats bar labels wide enough for the longest routec56d2ae
3. dashboard: stats totals as tiles, Unconfirmed as a flash result, a nameless device named by its kindd397ef4
4. dashboard: the stats page down to its numbersc40b38f
5. dashboard: the stats page is reached from the dashboard alone535fdf2
6. dashboard: the dev server refuses to serve the stats database550b9c7
7. dashboard: the stats page says when it can't load, wraps long names, and tells flash results apart5a33b27
8. dashboard: stats rate limits, a box active only when it connects, a device listed once two boxes share itaeea4d5
9. dashboard: stats headings named as nouns, a peak on each chart, and New this week among the totals848103a
10. dashboard: a volume for the usage stats databasef870b69
11. dashboard: a public stats page, and a note on the connect card that boxes are counted926b4be
12. dashboard: report each ROM download flash to the public stats1228dcc
13. dashboard: report each box, its cloned device and each update to the public stats72686a7
14. dashboard: a usage stats sink, and the totals read for the paged0c9d29
15. dashboard: an event route and a totals route for the usage stats04f1ad4
16. dashboard: usage counts in SQLite, a box known only by a keyed hash of its MACadd952c
17. dashboard: the stats server checks each reported event, field by field653dde4
18. dashboard: name ROM download on Advanced, and plainer copy for the fixed update pathd4a5e73
19. dashboard: a box the browser can't read says to replug USB2, instead of an empty liste6413eb
20. dashboard: say which firmware needs the resync before a handshake77af84e
21. dashboard: a flash retried after a pulled cable clears what the box still held, and the box's busy is no longer worded as a held port81471f1
22. dashboard: a box unplugged mid-flash reconnects when it comes back, and says the flash didn't finish57d455e
23. dashboard: a newer box is named as newer on every page, and Update leaves Advanced's result alone46e20f8
24. dashboard: Advanced keeps a flash set up when a box comes or goes, and spaces its fields like the other cards3f6a169
25. dashboard: one update at a time, and a newer box found after an install is connectedcb6c47d
26. tests: pin this page's half of the stable update pathd25cd11
27. docs, dashboard: name the stable update path, and tighten two comments1372324
28. dashboard: a refused file names no version to come, and Flash anyway has its own row3078029
29. docs: the update path keeps its shape, and Advanced's search words9aee6d0
30. dashboard: flash either chip or both over USB2 from Advancedc563c96
31. dashboard: an update records which slot each chip came back onf07950b
32. dashboard: say when a box is newer than the page, and where it can still be flashedcfd434a
33. dashboard: a box on another protocol is kept connected without asking it for health5713f0a
34. dashboard: a box on a newer protocol connects, for updatingbda9d27
35. dashboard: read a firmware file's chip and version before flashing it0e44693
36. dashboard: a USB device icon for a box64ffd79
37. dashboard: dropdowns no longer keep working in the background after they close2c3bca1
38. dashboard: pick an icon for each box in the sidebar3e36c5b
39. dashboard: long box names end in an ellipsis, and no Boxes section without Web Serial130805b
40. dashboard: Add box as the last box tab, and box card wording to match its siblingsa5285c6
41. dashboard: box list as sidebar tabs, Identify on the box card4c0664d
42. dashboard: one opener per port, and a box is held by one tabb3f5522
43. dashboard: every box on the PC, listed and held at oncecc61111
44. docs: an imperfect clone serves every IN endpoint2a7fdfd
45. docs: a lock is in force before anything sent after itf4d5828
46. docs: update firmware version strings to 3.4.40381ea2

Firmware

1. firmware, tools, docs: relayed reports reach the PC on the poll after they arrive, and none are lost across setting switches8ab665e
2. firmware, docs: a clip plays one entry to a report, every time2974980
3. docs: v3.4.3 in the release tablesdae9d77
4. firmware, tests, docs: a low-speed device's first report after a pause keeps its place557253f
5. firmware, tools, docs: a PC that stops reading a stream loses nothing, and reads it current afterd8f701e
6. firmware, tools, docs: zero-length packets on interrupt endpoints reach the PC as the device sends them1f918af
7. firmware, tools, docs: relayed endpoints keep every packet through halts, resets and setting changesebe9ed7
8. firmware, tools, docs: reports reach the PC sooner, and slower endpoints wait on its polls65aca8b
9. docs: the device shapes v3.4.3 clones, the OUT relay, endpoint halts and alternate settings08376dd
10. firmware: HID interfaces with alternate settings are cloned, and a switch loses no report7649b8a
11. firmware: an endpoint halt passes both ways between the PC and the devicee6d0427
12. firmware: every packet the PC writes to an OUT endpoint reaches the device, at the PC's rate41fdcee
13. firmware: IN endpoints numbered 5 and 6 are cloned72aaa87
14. firmware: a control request for more than 2048 bytes reaches the device as the PC wrote itf1d0046
15. firmware: a patched string is readable on a device that lists no languagebb769ca
16. firmware: the device qualifier and other-speed configuration are cloned7e160c8
17. firmware: a device with a small endpoint 0 enumerates at once8f6a057
18. firmware: strings are cloned as the device answers them, in every language it listsa1bf68b
19. firmware: a device with no HID interface is cloned7b63fb9
20. firmware, docs: remove the unused capture-owner hooks256946b
21. firmware: class and vendor requests to any interface reach the device45efb0e
22. firmware: vendor relay buffers are sized to the endpoints that can be live at oncefadb88b
23. firmware: a device past the host's descriptor tables clones what fits and says what did note3fa3d0
24. firmware: a device that refuses its own report descriptor is cloned refusing itcb82a3b
25. firmware: a slow report descriptor fetch is waited for, not left out of the clone3144bee
26. firmware: a HID descriptor after the endpoints, or listing another descriptor first, no longer loses its interfaced154a9f
27. firmware: devices whose interface numbers skip a value or start at 1 are cloned4f0ce37
28. firmware: an EP0 answer is relayed whole or refused4511feb
29. firmware: tell an interrupt transfer a zero-length packet endedabea180
30. firmware: count the IN endpoints a configuration can have live at once2ef987d
31. firmware: name the host limit a configuration exceedsf963527
32. firmware: one device is cloned from behind a hub, the first to configure50d7fb7
33. firmware: OUT credit that lets a clone hold the PC the way a busy device does82f7856
34. firmware: the report descriptor store keeps a device's refusal as an entry1a9a03b
35. firmware: vendor endpoint buffers are sized to what can be live at once93a9591
36. firmware: a table of string descriptors keyed by index and language41193f0
37. firmware: find a HID interface's report descriptor wherever its HID descriptor sits10e6ee1
38. tools: a descriptor gate and shared sim shape builders08d473b
39. docs: design and plan for faithful clones of more device shapes45a09a0
40. firmware, tools, docs: the box's reports reach the PC at one place in its framefad5327
41. firmware: moves at the poll rate stay one to a report beside a slower mousecffba9b
42. tools: the relay gate checks a feature report's bytes reach the device8d8aff7
43. firmware, tools, docs: a HID interface without an interrupt IN is clonedd88cfc3
44. firmware: rendered motion keeps pace with a flick (ABCurves 2.0)5c24af8
45. release v3.4.315efeb3
46. firmware: the host chip logs lost reports, not a heartbeat every 5 s3c69470

Library

1. capi: format a test31fd6ba
2. library, capi, python: relay_drops as the firmware counts it now that the box polls only as the PC reads47f1aff
3. library, capi, python: the zero-length packet flag on every relayed traffic class6516ce7
4. library, capi, python: relay_drops, RAW OUT and the device kind as the firmware has theme861041
5. library, capi, python: a clone is up once its identity says so, and the counters and kinds read as the firmware has them8d9ceb2
6. library, capi, python: a clone with no HID interface counts as cloned47df70c
7. library, capi, python: 3.4.3d3f2e13
8. examples: inject_latency also times a query's round trip4ce1787
9. library: a Windows read wakes on the box's reply, not on a 2 ms poll466412c

Docs

1. docs: relay drops, and the room RAW reports have in an interrupt IN queue15f09d3
2. dashboard, docs: zero-length packets on interrupt endpoints and the CATCH flag that marks them5f69d13
3. docs: relay drops, RAW OUT waiting, the requests CATCH sees on EP0, the product string and the device kindeb4619f
4. dashboard, docs: an attached device the box did not clone, relay drops, the IN endpoints a clone serves and the requests it passes on361a7a8
5. docs: the native requests page on a clone with no HID interface, and which long requests the box refuses9b2a754
6. dashboard, docs: a clone with no HID interface reads as cloned, and the IN endpoints a clone serves at onceca384a0
7. docs: when a native report keeps its own poll beside an injectionc721340
8. docs: a rendered report is bounded by the mouse's own field, not 127 countsb699bcf
9. docs: the library's unsafe code is its Windows comm-event wait8e996ba
10. docs: update firmware and crate version strings to 3.4.365632ec

Firmware

1. firmware: a render comment keeps the reason one report is outstanding1517fe2
2. firmware, tools, docs: fewer words in every comment and doc, and the claims the pass found wrong11752c1
3. docs: a flagged MOVE skips spreading and rendering with riding offe8dd0ca
4. docs: the protocol reference and architecture match the corrected behaviourf08625b
5. tools: gates for the advanced control fixes, session releases and OUT relay, all sim-driven68b5ebd
6. firmware: advanced control corrections, OUT relay fixes and report schedulingaafaf4f
7. docs: the version table names the flag RESET grew2e623a4
8. tools: the client's opcode list says what RESET carriesf3ad3a6
9. firmware: RESET requires its flags byte, like every other commandc7a4dd2
10. firmware: RESET can take the box's stored settings with itd69e608
11. firmware: a suspended PC is woken for injection, and hears it when it wakes10f89dd
12. tools: the render gate reads a property the paced fill cannot have1389ff4
13. firmware: the health bit answers for the control PC's motion, not the device's ownb39b53f
14. firmware, tools: comments say the one thing that is not obvious, in two lines8895c70
15. firmware: injected commands cross the link a whole frame at a time47cf5e8
16. tools: the wedge gate walks past an identity it has already spent3a4943b
17. firmware: an abort that worked no longer leaves the session open behind itca82234
18. tools: three cadence checks now read the value their property is about297b43d
19. firmware: what a departed control PC left behind is released whether or not a device is attached8aefc06
20. tools: the wedge gate stops spending a device identity it cannot get back91a3285
21. tools: a cadence gate was grading its own sender, not the box4298d03
22. firmware: an update no longer loses its answer to the link's own recovery3e1e82f
23. firmware: a relayed update says what happened to itb567d1a
24. firmware: an OUT report lost on the link is the relay shedding, not input the game missedfb52822
25. tools: the reference client's stats reply length follows the field it grew9e4111d
26. firmware: a drop counter says which fault it is, or it says nothing useful2af524b
27. tools: what a scale does to physical input no longer needs a hand2fca234
28. firmware: injected motion leaves on arrival, not on a clock of its own0ce9edb
29. tools: the transform gate measures the weigh and the transform together4960cd1
30. firmware: nothing that carries motion blocks on the link, and the host watches its own workfe925ec
31. firmware: the authority's answer crosses as the weigh and the release, apart4f02b05
32. firmware: the box reports what the links droppedc4bfad1
33. firmware: the motion authority releases the lock it takesabe6356
34. firmware: the motion authority moves to the host chip87f59c6
35. firmware: one weighing unit, on the delta rather than the reportefa315b
36. firmware: the motion primitives move to commondea849e
37. docs: the motion authority moves to the host chip9038058
38. release v3.4.26cdbbfe
39. tools: the conservation arm puts the sim back when it is done0154163
40. tools: the catch rig asks whether observing cost the player any movementac495eb
41. firmware: observing the traffic no longer costs the device's reportsff0c296

Library

1. library, capi, python: fewer words in every doc, and the claims the pass found wrong45468f3
2. library: a non-blocking control-port read avoids a WCH serial-driver crash on Windows5b60d0b
3. library, capi, python: the library restores its state after a box restart or session release37fdd95
4. library: RESET carries its flags byte on both paths460c51c
5. library, capi, python: the box can be asked to erase what it keeps across a reboote84f1b3
6. library, capi, python: comments say the one thing that is not obvious, in two lines6fc9f65
7. library, capi, python: back-pressure on a relayed stream is its own counterb5782b0
8. library, capi, python: control protocol 9 and the box's link drop counters57a56fd

Docs

1. dashboard: values that stand alone start with a capital47cdd60
2. dashboard: DTR and RTS reset neither chip, measured28c624a
3. docs, dashboard: fewer words on every page, and the wrong claims the pass turned up60d5c89
4. dashboard: shorter performance row labels33442a7
5. dashboard: the performance card shows every counter's counta196c66
6. dashboard: the relayed-streams chip stands on its ownf862c43
7. dashboard: decode protocol 9 catch flags and stats, and show patch and rewrite state as reportedda3a75a
8. docs: the library and bindings pages cover session recovery and the corrected advanced controlbbd45cf
9. docs: native pages for RAW, TRANSFER, REWRITE and PATCH, and a sidebar that fits its panef9c68d5
10. docs, dashboard: RESET carries its flags byte on every framed678a45
11. docs, dashboard: RESET's erase-and-reboot flagf231ce8
12. docs, dashboard: comments say the one thing that is not obvious, in two linesb648d75
13. docs, dashboard: the stats counters tell lost input from relayed back-pressure1cf6f21
14. docs: the stats offsets are what-inclusive, and the samples speak protocol 96bbb5f7
15. docs, dashboard: control protocol 9 and the inter-chip link counters37a8d9c

Firmware

1. docs: what protocol 8 changes, and the README's command referencef9bea34
2. firmware: control protocol 86388acc
3. docs: a packet trigger's read-back has its present bit clear3fd6757
4. docs: clip packet triggers44c550a
5. tools: the clip gate proves packet triggers on every surfacef219f0b
6. firmware: a matched packet is a clip trigger9ea7e94
7. docs: whole commands at the report rate14e9bf1
8. tools: the spread gate measures a report-rate loop, and a tool keeps the clone off the desktopd5c7ed6
9. firmware: a command loop at the report rate lands one command on each report807593e
10. docs: clip entries for pan, raw reports and transfers, and the clip transfer classd35751d
11. tools: the catch gate presents its own streaming mouse for the clock-offset leg9a73e43
12. tools: the reference client builds the new clip entries, and a clip gate proves them on the sim9bfa856
13. firmware: clips carry pan, raw reports and control transfersf2e6caf
14. release v3.4.1c09fa89
15. tools: the clone-policy gate waits out a multi-configuration capture45dd27e
16. tools: the descriptor-patch gate covers a device the box finds no mouse in66280e5
17. firmware: a descriptor patch is stored against the device, not the mouse in itd4aef98

Library

1. library: the C header names its ABI, and a reconnect refuses a box on another protocola0740f2
2. library: control protocol 8 and C ABI 8dac754e
3. library: packet triggers follow the box through the opt-in, and the mock refuses as the box does8208e81
4. capi, python: clip packet triggerseb0a3a7
5. library: a matched packet is a clip trigger7a93a44
6. docs: set_spread says what a loop at the report rate getsa006979
7. examples: the hardware suite proves clip entries and a clip across a reconnect901a1ec
8. library, capi, python: clip frames carry pan, raw reports and control transfers1a54225
9. ci: the crate, the C ABI and the Python binding must carry one versionc066099
10. library, capi, python: the bindings reach the transfer timeout the crate has9454600
11. release: bump the crate, C ABI, and Python binding to 3.4.1cecf373
12. library: raw sends without a round trip for the opt-in50a4f8d

Docs

1. docs, dashboard: what protocol 8 changes, the C ABI constant, and a reconnect across protocolsf85d02b
2. docs, dashboard: control protocol 8 and C ABI 89a0bf86
3. dashboard: number fields take whole numbers07ae27d
4. docs: packet trigger ordering, placement, timing and the mockb0b2f26
5. dashboard: the Clip card says when the box does not take a trigger439682f
6. docs, dashboard: clip packet triggers6bf5ab2
7. docs: what a loop at the native report rate gets from spreading3e4bb4e
8. dashboard: clips build pan, raw report and control transfer ticks8ac29c2
9. docs: clip frames with pan, raw reports and control transfers, and the clip transfer class52dce04
10. docs: the C and Python transfer timeout calls31a433c
11. docs: update crate version strings to 3.4.10a96dd5
12. docs: update firmware version strings to 3.4.1d8d5931
13. docs: a patch's cfg is a configuration index, and raw does not read the opt-ina0e2823

Firmware

1. docs: the transform section says how the two commands compose, not what one lacks4279cd1
2. firmware: a lock bites the field the device drove, and every one-bit class refuses a reversal5641e52
3. firmware: the two comments that still described the old pass orderde0c479
4. docs, tools: the lock scale is signed, and a transform only moves a fielddcfd81f
5. firmware: weighing a field belongs to LOCK, and a transform only moves it7af85dd
6. firmware: the field-transform pass is the first thing the renderer does, not the second5a28e90
7. firmware: a scaled swap banks each direction, and a remapped key is one a keyboard reports78b5873
8. firmware: the one injection path that reads the report map checks it is boundae091fd
9. tools: the transform gate reads the wire the game PC sees2322322
10. tools, protocol: the reference client and the byte reference cover transformsf90d0e7
11. firmware: a narrower configuration can still release what a wider one held2685068
12. firmware: transforms hold their own arithmetic, and a remapped key follows the button73941a4
13. firmware: shorter clip rendering notes and comments67f58ba
14. firmware: clips play through the renderer at their own pace, and rendered motion arrives whole74a6c6a
15. firmware: the mouse keeps working after the game PC restarts99e546d
16. firmware: the tools open the port at the rate the box answers on0bb18b2
17. firmware: a completed transmission is no longer thrown away before it is seencdcf9e0
18. firmware: a host image confirms itself only on a link that works4247979
19. firmware: the link's rate recovery starts where the image expects it52594cc
20. firmware: the inter-chip link runs at 20 Mbaudb7a799e
21. firmware: revert the short frame form and CTRL_MOVE839d5644
22. protocol: one version bump for the release, not twoe802ed3
23. firmware: the control link runs its own UART interrupt, and the link watchdog tolerates an erase9fab234
24. tools: the recovery flash script takes the control rate from the client, not a literalac67ce4
25. firmware: the inter-chip link runs its own UART interrupt instead of IDF's driver95d8c53
26. firmware: link frames go straight into the TX FIFO instead of through the driver's ring299a0a4
27. firmware: the snapshot retry budget is spent only on attempts the link could carry1bf86fd
28. protocol: the short frame form is v8, and the reference spells it out4d14a07
29. tools: a control-rate rig, and the client stops charging the port timeout to every reply1c5e7e6
30. firmware: the inter-chip link runs at 12 Mbaud and stops waiting out its receive timeoutab9f62b
31. firmware: a shorter hot-path frame, and injection stops paying for a refused emitb325bf4
32. firmware: the frame CRC reads a table instead of shifting eight times a bytebc14de0
33. firmware: hold 32 rewrite rules, and count the matches the reply reports5e96c48
34. tools: bearing and lock take the AC-Pan axis in their motion callbackf45dfef
35. firmware: descriptor-patch consistency read the wrong bytes, and a refusal stranded the clonef7426a3
36. firmware: rename the developer layer to the advanced control layer2e5b4a6
37. tools: catch subscribes to endpoints by number, and a raw injection gateecfc977
38. tools: reference client and protocol reference name an endpoint by number and directione9e99bf
39. firmware: name a cloned endpoint by number and direction, not a packed byte51f0e97
40. firmware: transforms apply on a secondary mouse, and a cross-class remap releases its key on removal3f136c8
41. tests: gate the field transforms and their compositions1481d38
42. firmware: field transforms (invert, scale, swap, remap)9c3a259
43. tools: reference client mirrors the widened input core (AC Pan, buttons past five)27cab02
44. firmware: correct AC Pan's clip scope and the reported button countf6741da
45. tests: gate AC Pan, buttons past five, and the spread/ride drain7052133
46. firmware: widen the input core to AC Pan and buttons past five5efc74d
47. firmware: the rewrite table's generation counter stays monotonic across a clear855f239
48. firmware: set the milestone version to 3.4.0617997d
49. firmware: skip an unchanged patch-set write, and correct a transfer status commente4dc8dc
50. tools: reference client and protocol reference track the patch-length widening2740192
51. firmware: correctness fixes to the developer layer's emit, observation, transfer and patch paths163e0f4
52. firmware: a table overwrite that will not fit keeps the existing entry7ee87de
53. docs: control-protocol reference for the developer layer (proto 7)8f0d7cf
54. tools: reference client for the raw, transfer, rewrite and patch verbs (proto 7)21817e3
55. firmware: dispatch the raw, transfer, rewrite and patch commands6d91f35
56. firmware: persisted descriptor patches, applied under the clone-refusal ladder50c48c2
57. firmware: health carries the rewrite state (health is now 16 bits)389aa18
58. firmware: an on-box rule table that rewrites traffic at every surface41e2c40
59. firmware: run a control transfer against the device on the control PC's behalfd17be06
60. firmware: raw report injection on any cloned endpointba45759
61. firmware: control-protocol wire for raw injection, transfers, rewrites and patches (proto 7)dbd264f
62. firmware: a per-device descriptor-patch set with consistency checksfc0a70f
63. firmware: a pure rule table for rewriting traffic in flight8e3e1cf
64. docs: implementation plan for the generic control layerd38d73c
65. docs: v3.4.0 design specs (generic control, widened input, transforms)ebaf3ff

Library

1. library, capi, python: docs say where to go, not what the API lacksa6f2520
2. library, capi, python: pin the signed scale's guards, and rank a block above a reversale2cbfde
3. library, capi, python: weighing a field is the lock's, and the percent is signed8856654
4. library, capi, python: the C transform op byte, and the tests that never compiledee62bcd
5. library, capi, python: a transform table replays in the order it was built8281209
6. library, capi: the clip ride readback fields keep their original wording837c25a
7. library, capi, python: clip riding reaches only the wheel while rendering takes the cursord37c615
8. library: say what actually confirms a firmware image7bfa7f4
9. library: the control link opens at 6 Mbaud6abe68c
10. library, capi: the rewrite table holds 32 rules38ca523
11. python: finish the advanced control layer rename in a test section header35160bd
12. library, capi, python: rename the developer layer to the advanced control layer2fa56be
13. library, capi, python: name a cloned endpoint by number and direction, not a packed byte42510de
14. capi, python: AC Pan and first-class transforms45d730e
15. library: field transforms (invert/scale/swap/remap)c085db7
16. library: a button blanket re-asserts every declared button after a reconnect7c4cebe
17. library: AC Pan axis and buttons past five in the input corea76de9a
18. library: C API and Python bindings for the developer layer0740836
19. library: serialize the developer-layer session state and harden its testsad99e91
20. library: cover the developer layer in tests and the hardware exampleab33cce
21. library: raw injection, control transfers, rewrite rules and descriptor patchesb492584

Docs

1. docs: a conformance gate, so shape drift is caught by the build and not by the readere7bf915
2. docs: an enum is documented on the enum page, not carded again on the method page67936ae
3. docs: pages point somewhere, instead of naming calls that were never there0bf11aa
4. docs, dashboard: the opcode table reaches 0x1E, and a block outranks a reversal2dcbfb2
5. docs, dashboard: the lock scale is signed, and a transform only moves a field412ff1d
6. docs: correct what the transform pages claim, and finish moving the types0fa563c
7. docs, dashboard: the v3.4.0 types live on the type pages, and transforms match the wireb822eb2
8. dashboard: the update choices are the ones they wereda23a7b
9. dashboard: reach a box on older firmware, and see its update througha949be4
10. docs: shorter clip rendering copy152ef34
11. docs: clips are rendered like live motion0c594e1
12. docs: the pages that tell you which baud to open still said the old one66182d3
13. docs: the control link is 6 Mbaud and the inter-chip link 209650135
14. docs: the rewrite table holds 32 rules02023b5
15. docs: rebuild the advanced control tab on the pattern the other tabs use12b760d
16. docs: gate the rewrite remove/clear buttons on the imperfect opt-in4fb4b97
17. docs: the advanced control dashboard tab is no longer blank without a deviced968e0e
18. docs: finish the advanced control layer rename (tab variable, a test comment)20213f3
19. docs: rename the developer layer to the advanced control layercc5d672
20. docs: fix the dashboard RAW comments left on the old packed wireffa5a43
21. docs: name a cloned endpoint by number and direction, not a packed byteb6988c9
22. docs: complete and de-slop the v3.4.0 reference7fbc25b
23. docs: AC Pan and buttons past five as first-class input22254a4
24. docs: widen the native HEALTH reference to the 16-bit wordda8f6eb
25. docs: developer layer reference and the 3.4.0 version bump9be03ac
26. docs: firmware 3.4.0, protocol 7 on the native pagesf70735d
27. dashboard: add a Developer tab to edit rewrite rules and descriptor patchesaec3079
28. dashboard: mirror the developer layer on the control link (proto 7)a4f09ec

Firmware

1. firmware: claim a deferred write before making it, not afterf4c4623
2. firmware: make a bulk sequence increment atomic now that it crosses coresea3e711
3. host: a control job refused rather than asserted before its queue exists7cbd464
4. firmware: move the two starved tasks instead of working around theme3be0b7
5. firmware: the record keeps the names worth reading, and a refused write is its own80566b6
6. firmware: watch fewer tasks, and say which one stoppede89e759
7. firmware: a busy box stays up, and a wedged one still names what stopped9aa62e8
8. release v3.3.487e90bf
9. firmware: a busy box no longer reboots itself under a heavy observation load16459f3
10. docs: what a watchdog record says, and the boot line as it is actually emitted6f56497
11. firmware: a task watchdog now says what was running when it fired698b263

Docs

1. docs: tx_drops counts both directions, not just the report path7318cb8
2. docs: update firmware version strings to 3.3.4ae7a32a

Firmware

1. firmware: ask for PSRAM a second time before giving up on itd69deec
2. firmware: boot a box whose PSRAM does not answer, and say what it foundc2a6425

Docs

1. docs: the firmware is on 3.3.26e79184

Firmware

1. tools: let the relay arm allow an accumulation, and only an accumulation881c34b
2. tools: let the mirror gate run itselfbcc50b7
3. firmware: keep the motion mirror true through a coalesced report52dec74
4. tools: three gate defects, one of which inverted what a gate measuredd1e6c1b
5. firmware: hold the smoother's rounding orderf61dfd9
6. firmware: deposit a rendered report from the clock that emits it9bcaf95
7. firmware: take a sixth off the model step and the observation passc52d7ef
8. firmware: learn the command period from the bucket's low edge42c5811
9. firmware: cover a command rate that varies continuously, not in stepsec76b3c
10. firmware: name the subject in three comments that opened on a pronounb2407d1
11. firmware: name what is being released instead of writing itee78ecd
12. firmware: keep the anchor a deposit has already earnedda65db8
13. firmware: drop the commentary the composing change made pointless22c6678
14. tools: wait for the clone's event node instead of guessing onea9d52c4
15. firmware: compose the release interval with movement riding4f77770
16. spec: record where the build differs from the design53697c9
17. firmware: fix the defects an adversarial audit of the release interval found4f52577
18. tools: gate the release interval on density and magnitude, not the total7b07f3c
19. firmware: spread an injected delta across the host's command interval5d8a740
20. firmware: release an injected pot across an interval rather than on one drain0eb0ee9
21. plan: the release-interval implementation7f03864
22. spec: spread injected motion across the learnt command intervalbc5d559
23. tools: three gate defects the hardware run exposeda31c648
24. firmware: say native for the operator's motion too, and drop two metaphors9f1de70
25. firmware: say native in the host tests too0f4bc36
26. firmware: say native9577f8e
27. firmware: fix the defects an adversarial audit of the render change foundcb5107c
28. docs: name the box state a faithfulness capture requiresdab36b4
29. docs: mark the superseded render design records1f57ef0
30. docs: correct the plan steps that contradicted their own constraints1ea2e65
31. tools: have the client's selftest read every option back125f3f1
32. firmware: dispatch and pack the box options in id ordere6588b4
33. tools: correct which personality the full-rendering gate runs on5e6c796
34. docs: record the rendered texture in the detection register6655022
35. tools: order the client's options by id, and name the bearing rather than the aimdfd08d2
36. docs: say rendering adds latency, and stop there250873a
37. firmware: credit ABCurves, and ship its licence with the images395771e
38. docs: cut the render prose back to what the surrounding pages carryf3b8c7e
39. docs: use the vocabulary and voice the rest of the stack already uses684f625
40. firmware: keep the wire version this branch already introducedfb83667
41. docs: record the soak result and what still needs a handba47579
42. docs: record what the acceptance run measured and what it did nota23b22e
43. device: keep the operator's motion out of every path that drops the host's3169ec7
44. docs: describe the texture as its own setting, and what full rendersa74ee47
45. tools: address the texture settings on their own command19da53d
46. device: render the attached device's motion with the same model483084c
47. docs: plan the work for rendering native motion7af1de1
48. docs: settle how native motion joins the model's streama4cc658
49. docs: the constraint set and open decisions for rendering native motion58ad3d3
50. docs: design for spreading owed injection across the learnt command interval61bfe18
51. tools: measure what a host commanding slower than the device costs the wiref8fa8ec
52. docs: describe the kernel the box ships and the clock the model steps on5ea77cd
53. docs: record every departure from the upstream render runtime and its measured cost5f5d897
54. tools: verify the profile armed before the texture page claims a mode6635fda
55. tools: apply the comparison kernels causally, at their own lagbcc39ba
56. tools: resolve the clone's input node instead of assuming its index6b571ae
57. firmware: begin a render phase on the tick it steps on, and hand back what it still owes78545fa
58. firmware: render the intent in single precision, on a kernel that answers soonerc0c16c4
59. tools: ignore the texture rig's outputcc46e41
60. tools: measure the render model against injection shapes a host actually sendsb2165e2
61. tools: score injected texture against the attached mouse9a513ab
62. firmware: render on the same clock reports leave on95e6e95
63. docs: record the measured result and why the render path deposits againa43c4a8
64. tests: bound the assertions that passed on a dead pipeline6925be7
65. firmware: drop the smoother's dead branches and answer flush misuse78419a3
66. firmware: keep the device stack on its own core, and let the pace clock catch uped1d30b
67. docs: correct the renderer state size after the smoother change4ee0cab
68. docs: what the render model reproduces, and why it paces off a timer8286116
69. tools: gate the renderer on its texture and on whole delivery, not on idle fractionf978169
70. firmware: stop the render pace clock on a clear, and only yield when a tick is already owed43978f4
71. docs: record the measured injected-emit latency41ee33c
72. tools: measure injected-emit latency per render modef3e1a14
73. firmware: the renderer keeps to its millisecond grid and starts a phase without folding the profile94efd3c
74. firmware: injected motion emits as soon as its rate gate allows8fb49d3
75. firmware: one place that builds a synthetic injected frameac55a88
76. tests: assert the render pipeline's latency without hardware9eed0f9
77. firmware: the de-spiked smoother no longer drops the motion it starts with383cbb2
78. docs: implementation plan for the injected-emit latency workf0cb115
79. docs: design for cutting injected-emit latency6d03dae
80. docs: use one spelling for the words that had twob700ceb
81. firmware: apply a render mode change at the next phase, not mid-phase03afab9
82. docs: correct stale protocol facts and tighten the writing07fdee7
83. docs: split the render section and drop the em dashes75735d7
84. docs: correct the protocol version and the emit sectiondf9d435
85. tools: assert the render profile is armed, and leave the box as found6212b28
86. firmware: render only once the device has been seen to movefcfbef5
87. firmware: render injected motion by default092b7f7
88. firmware: give the render mode its own field in the emit optionef63132
89. tools: show every render mode in the texture page and fix its idle metric2ad2995
90. firmware: stock / de-spiked / off modes for the injected-motion smoother2b5bfd2
91. firmware: no first-tick spike from the injected-motion smoothere0fc07a
92. tools: ignore the generated render_viz.htmlc3bffde
93. tools: visualise paced vs rendered injectionfd2550d
94. firmware: drop the stale mode-3 name from the render headerab2e69c
95. firmware: make rendered a texture flag that composes with the pace55622e2
96. firmware: vendor the ABCurves MIT license2b8f7a4
97. firmware: cut comment noise in the render path0f30828
98. release v3.3.1edcb98f
99. docs: list the rendered emit gate among the validators3561e40
100. firmware: measure the rendered emit gate in isolation, off kernel time482d940
101. firmware: harden the rendered emit path found by reviewd6837e4
102. firmware: point the render gate at the trace mouse's own product idfdbb35f
103. firmware: conformance gate for the rendered emit mode7fa49cc
104. firmware: keep the rendered tick on-chip (IRAM code, internal-RAM tables)183370b
105. firmware: rendered emit mode drives injected motion36ec37a
106. firmware: move the clip ring to PSRAM to free internal RAM70cec1c
107. firmware: 1ms grid reconstruction for the emit observera5c2961
108. firmware: bit-exact render runtime with SIMD-optimised tick and host gatef9a10b3
109. docs: implementation plan for rendered emit modebbd6c1d
110. docs: design for rendered emit modee9407bb

Library

1. library: name what is being released instead of writing it2cfd1df
2. library: say native for the report rate in set_spread's docs4f86ba2
3. library: regenerate the C header, and widen two boundary tests past a bytec8e0104
4. library: expose the release interval on the library and both bindingsef0d6e6
5. library: say nativee047fee
6. library: cover the render option's async and C-ABI paths, and match the mock to the box8a74375
7. library: expose the box options in id order across the library and both bindingsfeadf82
8. docs: say rendering adds latency, and stop thereb76c1d0
9. docs: name the model the box renders withbad84cd
10. docs: tighten the render option's doc commente64ec71
11. docs: use the vocabulary and voice the rest of the stack already uses788be0b
12. library: keep the wire version this branch already introducedb1b27dc
13. library: make the mock answer the pace the way an unarmed box does11f7b86
14. library: address the texture settings on their own command5f31ad4
15. library: one spelling in the Cargo.toml comment6b1f0b5
16. library: take the render mode across the C ABI as a bytee3a04f2
17. library: correct stale doc comments and tighten the writinge2ff4f9
18. library: bump the C ABI, and require an explicit render modef997d6c
19. library: the render mode is its own emit field, and de-spiked is the default860905c
20. library: a RenderMode of off / stock / de-spiked / unsmoothed4fcb605
21. library: trim the emit-pace doc comments to one-liners1971cf3
22. library: rendered is a composing flag, not a pace variant9787636
23. release: bump the crate, C ABI, and Python binding to 3.3.15dbfc7d
24. library: assert the rendered pace resolves to 1 kHz, and document it59965e5
25. library: rendered emit pace across the crate, C API and Pythone5014c9

Docs

1. docs: name what is being released instead of writing it23ad387
2. docs: drop the riding paragraphs the composing change made pointlessce9cc43
3. docs: say that the release interval composes with movement ridingbc1453c
4. docs: correct what the release interval does, and say native02d005b
5. docs: document the release interval and add its control1dfa40e
6. dashboard: say what two more controls are, in fewer words50db963
7. dashboard: say native for the wire rate toob531d09
8. docs: say native068eab6
9. dashboard: name the render setting, not what it does to the mouse64cc05c
10. docs: fix the docs defects an adversarial audit found611014b
11. dashboard: order the option code paths by id too82b1bec
12. docs: list the box options in id order on every page3bf7bb0
13. docs: put the option cards back in the order the other pages usebcc5402
14. docs: say rendering adds latency, and stop theree7b78ef
15. docs: credit ABCurves where the render option is described861e177
16. docs: bring the render copy back to the shape of the cards around ite57335d
17. docs: put the render option back with the other optionsd24381c
18. docs: use the vocabulary and voice the rest of the stack already uses8f43a93
19. docs: give the texture its own page instead of a footnote in the options list9c60f5f
20. docs: keep the wire version this branch already introduced53210af
21. docs: correct the texture's own cost and what it renders when it is off5732204
22. docs: give the texture settings their own command and page8145b5c
23. docs: shorten the labels that had grown into sentences76b9167
24. dashboard: give every card a search anchor and every state its own card6cffdd2
25. docs: correct stale references and tighten the writing945600a
26. docs: cut the emit card down and stop tables repeating their own label17cb075
27. dashboard: say what to do, not what the protocol isb0b2132
28. dashboard: give the older-protocol state a card, not a loose callout285ddfe
29. dashboard: connect a box a protocol version behind so it can be updated491ed70
30. docs: correct the protocol version and the binding references5051cc4
31. docs: rebuild the emit option page around its four fieldsbbaa865
32. dashboard: four render modes in the selector; drop the Render Lab7a483cf
33. dashboard: hint when rendered emits nothing on an unprofiled devicebc163c1
34. dashboard: a Render Lab to visualise rendered vs paced injectiona712679
35. docs: fix the rendered control's label case and reactive blurb5317a5c
36. docs: strip the rendered prose to the terse house styleb5cb437
37. docs: match the bindings set_emit_pace signature and status fielddd7db68
38. docs: rendered is a composing flag, and credit ABCurves34932ad
39. docs: order the option sections by id8b77755
40. docs: tighten the rendered emit prose3a70975
41. docs: rendered emit mode, and the 3.3.1 firmware versiona090bcc

Firmware

1. release v3.2.1c024681
2. docs: what the box does when a mouse stops answering, and what the delivery count meansc2c87e3
3. firmware: count a delivery stall only on an endpoint the PC reads8c48083
4. firmware: bring a stream back when the mouse stops answering its own endpointaa409d1

Docs

1. docs: update firmware version strings to 3.2.17e3ae55

Firmware

1. docs: correct the setup design to what the board and the reviews showed8ff0f49
2. docs: the build order for getting a box set up the first time05e0635
3. docs: the design for getting a box set up the first time29e6813
4. tools: guard each restore, hand the sim back, and name the verb that existsbd6bbde
5. docs: correct the forced-rate entry to what the wire actually does16645c1
6. firmware: build the forced rate the way the files around it are built346c220
7. firmware: say the poll interval changed only when it didf4b8491
8. tools: run the rate gate against a real attached device, forcing away from what it declares696c824
9. tools: let the riding gate run without a hand when the sim owns the only node45e3456
10. tools: build the gate's mouse from the rig's own, and find its input node77e7a57
11. firmware: what the review found in the forced-rate path48d477d
12. tools: a gate for a clone running at a rate the device did not declare25ff60e
13. docs: how a box is told to run at a rate the device did not declareafbb2a0
14. tools: set and read the forced rate from the reference client0527fda
15. firmware: ask the device for reports as often as the clone says it will be asked94fe1f7
16. firmware: let a box advertise a rate the device did not declarec99497e
17. firmware: carry a forced rate and what the clone advertises on the pacing optionba558cc
18. firmware: work out the poll interval a forced rate needs, and write it into a descriptor4c97cb3
19. docs: keep the dash sweep out of the file it sweepsa9d859e
20. docs: the build order for a clone that runs at a forced ratee9b5d91
21. docs: the design for running a clone at a rate the device does not declare068a5a2
22. gate: hold each chip to the window it advertises43ecf6e
23. firmware: let the mouse-side chip restart the way it always didd36c59b
24. firmware: put the mouse-side USB controller down before that chip restarts6165e95
25. firmware: ask for only as much as the chip can actually hold8f1efa9
26. firmware: hold the whole window the box says it will takec9e5a55
27. docs: say that abandoning a commit disarms both chips4965827
28. firmware: abandoning one commit abandons both, and the client could not clear either493bb1a
29. firmware: tell the other chip too, and settle before erasingce01bc4
30. firmware: only abandon an update while there is still something to abandon898ff28
31. firmware: what the third review found in the second round of fixesa1a3075
32. firmware: what the second review found in the first round of fixes0bba54a
33. firmware: anchor every timer to the clock, not to zero6f6728b
34. docs: match the renamed client callc08bba0
35. firmware: read every elapsed-time check as signedc07b370
36. tools: what the review found in the reference clientd03ca93
37. firmware: what the review found in the update pathac065b0
38. tools: one command to push firmware to a connected box211c39f
39. docs: how an image reaches each chip now4711a91
40. tools: prove a broken image does not stick20965e3
41. tools: a gate for taking an image over the control porta9f26b8
42. firmware: ask the far chip for its version until it answers1da1093
43. docs: record what the hardware corrected in the update design9dbabfc
44. firmware: keep the clone down while the box rewrites itself20a4807
45. firmware: serve an update on the control port and pass one along0a1d078
46. firmware: let the mouse-side chip take an image over the internal linka37c3b2
47. firmware: put the box down cleanly before it rewrites itself161bbce
48. firmware: write a staged image to the spare slot1f31d89
49. firmware: the chunk sequencing behind a staged image, with its testsa139518
50. firmware: the wire shape for taking an image over the control port1afe07e
51. firmware: give both chips a second app slota22f3c1
52. docs: the host chip gets one rom flash, and it has to be the right onee14b65a
53. docs: the work order for updating over the control portde9c107
54. docs: the design for updating both chips over the control portedbd2ef
55. docs: the G502 answer is not random, it is before-first-motion8944b4c
56. docs: point the work order at what actually shippedb36bd88
57. docs: record what the review corrected in the at-rest rule88c9289
58. docs: say which fields the spec pins and which only look pinnedcb4cd80
59. firmware: trust only what the descriptor pins, and ask once at a device that answersbc6d4ee
60. docs: where an at-rest report comes from, in order of preference80bef6d
61. firmware: ask a device once, and bring it back if the question stops it answeringe694922
62. tools: a gate for a device that stops answering when askedfe39acb
63. firmware: reuse a report the box already has before asking for one7f8e023
64. firmware: do not ask a device for a report its descriptor already spells outf92d3b6
65. firmware: work out from the descriptor what a report reads at resta3c6f8e
66. docs: the build order for deciding when a device needs askingd88e6e5
67. docs: decide from the descriptor whether a device needs asking at alle0b4c26
68. docs: why a receiver falls into its bootloader behind the box6cfccd4
69. fix: a full key bitmap made a receiver's paired mouse report as a keyboard03f5d3e
70. docs: name the mechanism in the protocol reference and the comments5fc4378
71. docs: correct what a discard, a refusal and a saturating gain actually doccf4f60
72. tools: validate the motion flags, and pin what the last round fixed282386b
73. tools: stop a mutating verb defaulting to the destructive value6ea1bb0
74. firmware: make a bearing clear actually clear, and keep the aim's angle0e59085
75. docs: write down what the box applies, not what it was told85adde4
76. tools: make a bearing gate that a plain sign lock cannot pass252e352
77. tests: pin the weighing rules a mutation can still break32603dc
78. firmware: weigh a report only when the scale actually changes it13d1106
79. tools: stop the puller before closing the port it writes to698a65f
80. tools: measure what the bearing does to a hand, not what it reports125d7c2
81. firmware: let a button store the block it renders, not the number sent0895364
82. firmware: re-read the version header when it changes, not once per build tree6c56608
83. tools: validate a scale by hand, not only a lock9a15813
84. firmware: let a lock keep a fraction of an input instead of all or none19606fe

Library

1. ci: check the features this crate actually has, and do not ship wheels for a crate that did not publish5ed374b
2. library: mirror the interval the box now advertises, and format66f588e
3. library: what the review found in the forced-rate surfacesee75bd6
4. library: check the forced rate round-trips on a real box06b2435
5. bindings: carry the forced rate through the C and Python surfacesd31e06b
6. library: set and read the rate a box advertises94d2cec
7. library: shorter refusal textbc78c27
8. library: say what a refused update means and what to do about it5cb91b6
9. library: derive the result-name guard instead of listing itcdca871
10. library: clear what a refused commit left armed0a6dc24
11. library: drop a reply the box has moved on from74b6e09
12. library: look for a parked reply on every wake5606b85
13. library: park a reply for the caller it belongs tocc701a1
14. library: put back a reply that answers another caller2df1b87
15. library: what the review found in the update client0b7cafd
16. library: formattingee4567d
17. library: update box firmware over the control portb2b43f6
18. docs: name the mechanism in the public API surfacef1fdc4b
19. capi: refuse a byte no constant names, everywhere one crosses34747d2
20. hw_full: assert what only the firmware can satisfycb8f075
21. bindings(python): check a scalar before ctypes truncates it8826d5a
22. capi: take a byte where the boundary cannot trust an enumd80ce4b
23. crate: model the box's lock table so a reapply rebuilds what it found2c98fac
24. crate: say what the scale readback actually promisesd2574c3
25. hw_full: prove the scale round trip and the bearing option on a box78db22f
26. v3.2.0da8d355
27. bindings(python): follow the C ABI to the scale system8bb2c5b
28. capi: carry the scale and the bearing across the C boundaryed65e1d
29. crate: let a lock keep a fraction of an input instead of all or nonee179bcb

Docs

1. dashboard: let the manual flasher be manual, and size the secondary buttons like the reste643106
2. dashboard: keep the step chip, and stop the diagram forcing the gap above it54d3a70
3. dashboard: put the step in the header instead of a row of its own041af30
4. dashboard: even the gap either side of the step counterc769401
5. dashboard: let the card header own the step counter's spacing0a21b47
6. dashboard: give the step chip its own gap, and say the reason oncef8851e0
7. dashboard: set the device port apart from the two computers86fa2a2
8. dashboard: one diagram per action, and cut the filler4ed10e7
9. dashboard: judge each chip on its own version, and tests that can actually fail324cc80
10. dashboard: a finish screen that could not finish, and the update path nothing tested7996f79
11. dashboard: stop claiming a check that never ran, and cover the page that had no tests5a29691
12. dashboard: two fixes that cancelled each other, and four more blind tests1c4c470
13. dashboard: the wrong chip named, the slow read, and three blind tests397baba
14. dashboard: the button beside the socket, and what round four foundd224f71
15. dashboard: the page freezes and the skipped gate that round three foundc0ec1ef
16. dashboard: what reviewing the fixes found in themce8f568
17. dashboard: what three reviews found in the setup pathde0efd7
18. dashboard: a guided first install, and a connect that says what is wrong9992bc6
19. dashboard: turn a failed connect into one answer, and say which firmware gave it0c952e7
20. dashboard: name the wire rate in the card's own heading stylecac69b8
21. dashboard: build the wire rate the way every other control on the card is builtc078168
22. dashboard: give the wire rate the card's own spacing, and name the two rate fields apart260a4e9
23. dashboard: apply the forced rate the user actually asked for55b2243
24. docs: the rate a box advertises is now something you setfbfe5de
25. dashboard: refusal text that reads like the rest of the UIa53ca34
26. dashboard: errors that say what to dof40db50
27. dashboard: adopt a port that is already open, and say the window is not fixede108e85
28. docs: write the update pages the way every other page is written7d6fd66
29. docs: put the update feature where the rest of the docs keep things42efafd
30. docs: say that abandoning a commit disarms both chips86c2a10
31. dashboard: make the typecheck gate able to fail88f8787
32. docs: list the new result where the results are listed8a7bb6c
33. dashboard: typecheck what the bundler never looked atc835fea
34. dashboard: clear what a failed update left behindaa668f7
35. dashboard: keep the real cause when a port goes away24c8e42
36. dashboard: do not consume a reply from an abandoned transfera2f10f8
37. dashboard: do not lose an update reply that arrives earlyb9d383a
38. docs: say which flashing path is the normal one now009a9a4
39. docs: the library calls for updating a box2209088
40. docs: a page for updating a box over the control portb410928
41. dashboard: update both chips over the port you are already connected to9cfdd68
42. dashboard: speak the firmware-update protocol2f2deb1
43. docs: say where a lock readback can disagree with what the axis doese30951b
44. dashboard: keep a readout's rows together64f7075
45. dashboard: give the riding checkbox the margin its siblings have33ccc30
46. Revert "dashboard: let the container own vertical rhythm, not the child"ed9526e
47. dashboard: let the container own vertical rhythm, not the child99c812d
48. dashboard: stop restating the reference in the control panel2361d38
49. dashboard: pin that the release-on-silence fact is stated once629aeba
50. dashboard: one sentence per control on the control page too9ecc171
51. dashboard: put the bearing Turn off button backd67a9e1
52. dashboard: one sentence per option, and one shape for every control58977c9
53. dashboard: say why a connect failed on the control page3690611
54. docs: name the mechanism, not the picture it makesc5bfc00
55. docs: hold the rest of the reference to the bar the lock page setdc0016f
56. docs: rebuild the lock page in the shape the rest of the reference uses5a46100
57. docs: correct what a saturating gain does, and make the client claims truebc998d7
58. docs: split the lock page, and say what the bearing actually does660454e
59. docs: say each thing once, where the reader is looking for it2aa0d86
60. docs: carry the scale, the aim and the version the pages describe45169ab
61. dashboard: drop a lock entry the box would never have named7ef46d8
62. docs: correct what Both does, and warn where the version cannotef59816
63. docs: document the scale and the bearing across every page they reachfbf8811
64. dashboard: put the scale and the aim bearing on the control page80bbf96
65. dashboard: follow the lock wire from a state to a scale3bf110d

Firmware

1. docs: the design behind the per-command riding override4aa8246
2. tools: a gate for what riding delivers, and what it holds back4598c8b
3. tools: two tools had been sending move frames the box throws away269decb
4. firmware: let one move say whether it waits for the user to moved9da046
5. docs: say when a clip trigger takes its input lock, and when it does note3790a7
6. firmware: lock every axis when the blanket asks for itc8598cd
7. docs: write the catch design doc without the dashes3ce7a00
8. docs: the design behind the catch API rework5b82fe9
9. firmware: charge a lost event to every subscription it belonged tof691b7b
10. tools: leave the rig on a device the other gates can actually exercise54750b7
11. firmware: make a snapshot say which class and which edge produced itc2f4760
12. firmware: hold back only what can flood the link, and let the waits be short enough to matter375267f
13. firmware: make the clock estimate say what it actually knows, and the tests able to tell003518e
14. docs: say what the counter orders and what actually holds input earlye791ae4
15. firmware: keep the observation stream from making an input report latef5ab3a7
16. tools: a conformance gate for the observation surface6eee30a
17. firmware: keep the busiest observation work off the core the USB stack needs1877297
18. firmware: carry the crash record's caller chain, and keep the rig off the desktop cursor52c70ad
19. firmware: wake the drain directly instead of through a shared counterbcdea80
20. firmware: read the observation table without locking the paths that read it mostcce63f1
21. firmware: pace the event stream to the link instead of spinning on it8219d3d
22. firmware: keep the event stream from stalling the link it shares255569e
23. docs: document what can be observed and which clock stamped it0f132a6
24. tools: prove observing the traffic does not disturb it39fb9c0
25. firmware: observe every class of traffic the box carriesbfbed83
26. firmware: hold the observation subscription as a table, and rank its event streamcb46f50
27. firmware: stamp bulk where it arrives, like every other inbound stream7c8ca4b
28. firmware: a four-stamp exchange so the two chips' clocks can be compared619e26f
29. tools: teach the reference client the new observation surface9694782
30. firmware: measure the two chips' clock difference instead of assuming it380864b
31. firmware: address what to observe the way locks are addressedc7c200c
32. docs: the implementation plan for observing everything the box carries52d3182
33. docs: tap emitted reports where they complete, not where they are staged59c8542
34. docs: a design for observing everything the box carries, not only physical input96ee0f4
35. docs: a top-level guide so the repo carries its own bring-up knowledge6bb87dc
36. tools: prove the descriptor half of the configuration cost, on hardware11b67c1
37. firmware: cost the configurations the clone is not in, not just the one it isb23f831
38. firmware: the faults only repeated switching could reachbc626ca
39. firmware: rebind the roles a configuration switch moves, and hold every configuration's descriptors8092fd6
40. firmware: a configuration switch rebuilds what it changes3d72c39
41. firmware: the rest of what the reviews founde346dc6
42. firmware: a marker that is a prefix of another could never be seen missing3c5215e
43. firmware: defects a review found, including two the previous round introduced368fb59
44. firmware: carry the report descriptors a configuration switch needs, with the switch76e3a12
45. firmware: enter any configuration a device offers, not only the one it came up inb262349
46. firmware: carry a control data stage larger than one link frame62b41b1
47. tools: count the bytes on every endpoint a shape exists to exercise7a7c042
48. tools: make the hardware oracles able to fail2853b1d
49. firmware: defects found reviewing the rest of the interface-coverage workfcfbd14
50. firmware: defects found reviewing the control-endpoint work7050f85
51. firmware: carry a control endpoint a device declares beyond the firstdf18d25
52. firmware: name the control endpoint above zero, and teach the port to hear ita1c527d
53. firmware: describe every configuration a device has, not just the one it is incccaa2d
54. firmware: follow the alternate setting a host selects, all the way to the device51bb6d0
55. docs: why the alt-setting relay is asynchronouscd6ed02
56. firmware: drive the endpoints a HID interface declares and its driver ignoresf8a0d11
57. firmware: carry a report descriptor the device actually has, from an interface of any classab95091
58. docs: design for retiring every refusal the hardware does not forceb29edaa
59. firmware: a control transfer too big to buffer hung the clone instead of failing it66580fc
60. firmware: the two chips disagreed about which interfaces they serveeba56a3
61. tools: the refusal ladder had no test that ran itc9f688a
62. firmware: an audio device cloned in silence, minus the half that matteredae6014e
63. firmware: a HID interface could declare an endpoint the clone never feeds866ff13
64. firmware: a clone answered questions about configurations it does not have90e3d13
65. docs: the vendor bulk ceiling is measured now, not estimated7814fb3
66. docs: the vendor-interface pass-through design0494a35
67. firmware: vendor-specific interfaces were declared to the PC but carried nothing446b93f
68. firmware: a rejected hunk could still report itself applied9ece3cc
69. firmware: defects found reviewing this branchd0a5751
70. docs: the low-speed frame-interval fault, and finish a truncated sentence5584bbe
71. firmware: low-speed devices were told the bus was idle on every frame6d28b89
72. firmware: one keypress read as three on a keyboard that splits its keys across reportsc3c4aa3
73. firmware: the previous commit changed no behaviour; say so82d9c02
74. firmware: log which firmware image is actually runninge5add6f
75. fix: device kind was decided by interface order, not by what the device sendsb91235b
76. fix: queue overflow dropped the newest report on one of two paths5459acc
77. fix: report-ID budget counted Feature reports it never neededa38482c
78. Revert "surface reports cut short mid-transfer"bedda73
79. fix: key reports corrupted on an endpoint shared with a mouse reportcf55d44
80. docs: low-speed clones now match native bus speed2d0077f
81. tools: screen HID devices for multi-packet reportsb0e1a84
82. fix: low-speed device support on both chipsac5a647
83. docs: compress the protocol timestamp note4195634
84. docs: spec matches the wire-width crate surface7224a5d
85. docs: usage-event timestamp capture from the bench694a185
86. tools: revive the lock validator and poll for ROM download20f668d
87. docs: native jitter baseline for the timestamp accuracy oracle3f73ce6
88. docs: record the bench device's measured report families3d31dbb
89. firmware: catch events carry the report's arrival timestamp0722057
90. docs: timestamped catch events design spec644ea05
91. revert: drop the QUERY(LINK) selector283e47c
92. fix: defects found reviewing the control-link work4a2962f
93. firmware: drop the emit-path blocking histogram from RESP(LINK)6307db2
94. firmware: RESP(LINK) reports only the window it can measure exactly8974cf2
95. v3.1.00b4e07e
96. tools: control-link loss rig, scripted host flash5e0cc83
97. firmware: control frames survive a busy emit path3a1a72b

Library

1. bindings: carry the move override through C and Pythone0975ce
2. crate: let one move say whether it waits for the user to move35034c2
3. docs: strip the dashes and the padding out of the referenceecfab01
4. crate: address an input the way a lock does, and hand back the edgesc6f5ca0
5. crate: refuse a subscription the box cannot hold, and send only what changedb341abf
6. hw_full: measure the sustained rate, and drive the traffic a still device will not39cc619
7. crate: route a held-usage snapshot as the class state it is, and mirror it in C and Pythonce3a72e
8. crate: expose the protocol version, and stop asserting quiet on a streaming clone8ee86f3
9. hw_full: check the per-id observation path on hardware, not only the blankets344230a
10. crate: route each observed input event by the address it actually carries3cfb471
11. crate: give each observer only the traffic it asked forc6ea88c
12. bindings(python): follow the C ABI to version 46bbe0cf
13. capi: carry the observation surface across the C boundary911eb2f
14. crate: observe every traffic class, not only physical inputcd50aca
15. capi: hold the ABI version at 37ce742b
16. v3.1.0956d9c1
17. refactor: hand the catch timestamp over at its wire width8b642b5
18. refactor: match the crate's idiom in the catch clock widener72e76b1
19. feat: catch events carry the report's arrival timestamp81118e6
20. v3.0.1f6b4dd1
21. fix: stop Windows injection queueing behind the reader thread7ba7c4c

Docs

1. docs: check the rest of the pages this feature touched the same way6f6d350
2. docs+dashboard: follow the patterns the reference already has, on the pages this feature touchedfdc3b21
3. docs: document what a move can say about waiting9bb4f27
4. dashboard: drive the riding override from the control pagea3e52c4
5. docs: index the names the catch rework introduced3e2ed8b
6. docs: index the dashboard cards search could not reach737249f
7. dashboard: name the clip trigger flag in full on the checkboxa59bf55
8. dashboard: call the clip trigger flag what the protocol calls ita60a88e
9. dashboard: open the clip trigger picker on the class its own list leads with96b1eca
10. docs: describe a clip trigger's input lock by what it does5f03e36
11. dashboard: rebuild the control page around what the box can do08e613c
12. docs: cut the documentation down to what it actually saysf4d9e68
13. docs: stop the reference justifying what the table just definedb7a1698
14. docs: stop the tables saying their own column headings2753832
15. docs+dashboard: document the catch surface as it is, not as it wasf43090c
16. docs: a table that is full, and who a lost event is charged toa1a4b14
17. docs+dashboard: say which edge produced a snapshot, and how to route one8f9f0e0
18. docs: document the protocol version a build speaks5ee0d93
19. docs+dashboard: say which class a snapshot names, and what an unfitted drift rate meansee104d3
20. docs: correct what the event counter means and how the stream is rankedd43d7bb
21. docs+dashboard: document and display every traffic class the box can observe1bad5f4
22. docs: repoint two search entries at the anchors that existe3b9ca8
23. docs: ABI displays back to 381c9e65
24. docs: crate and ABI displays for v3.1.055387ed
25. docs: firmware displays to 3.1.04ef0046
26. dashboard: decode the catch timestamp instead of reading it as motion90cc283
27. docs: ts_us belongs on the frame that carries it37146b0
28. docs: trim the timestamp cards to reference-only3026e95
29. docs: rebuild the timestamp cards as diagrams and tablesda0dc25
30. docs: correct the mock inbound-traffic card subtitle1b2c016
31. docs: catch events carry the report's arrival timestampbd40044
32. docs: crate version displays to v3.0.1981a99e

Firmware

1. firmware: classify falls open on report shapes it can't model73d0dbd
2. firmware: a locked input emits nothing instead of an empty report (v3.0.1)7abc2a8

Docs

1. docs: firmware version displays to v3.0.19f6257b

Firmware

1. docs: update README control-surface to proto 3 (MOTION_EVENT/USAGE_EVENT, CLIP_SET/CLIP_TRIGGER, 10 query selectors)d495261
2. ci: split the Discord release changelog across messages over the embed limitb26bdea
3. docs: control-protocol clip trigger set + config + RESP(CLIP) rewrite8d02390
4. tools: refclient clip trigger set + config, new RESP(CLIP) decode3d2ca96
5. firmware: clip trigger set, retain/rewind engine, and queryable configfeefbe2
6. docs: clip trigger + config redesign spec and planc872c0c
7. cleanup: strip comments and AI-slop from host firmware, common, and docs4bde84e
8. cleanup: strip AI-slop and comments (tools, host tests, protocol doc)26a0a08
9. cleanup: strip firmware comments and AI-slop9eaea0b
10. medius.py: one usage vocabulary for inject and lock8027a85
11. docs+comments: proto 3 wire spec, RESP(CATCH) media bit, stale event-name comments5f1f6dc
12. v3.0.0e939e75
13. medius.py: decode the clip held-usage snapshot in RESP(CLIP)b1b60ca
14. firmware: report clip held state as one usage snapshot, not a button bitmask0321ece
15. firmware: one catch event shape for every input class9b79c45
16. firmware: uniform LOCK addressing and a full lock-state readback (proto v3)d3d536b
17. firmware: one inject verb and one synthetic-emit path for every class093d3f8
18. firmware: unify the HID injection core on one momentary-usage model30e27fd
19. docs: input-taxonomy unification design8dc8f7d
20. firmware: field-generic clip control with an autolock scopef977fb5
21. firmware: clip STOP releases held input; auto-lock owns locks per directionf76849e
22. firmware: clip auto-lock locks all input, drop the lock-mask3ad676c
23. firmware: buffered clip playback drives keyboard and media, not just the mouse255ebd3
24. docs: note buffered clip playback in README, ARCHITECTURE, tools2c2a9c7
25. docs: correct clip overrun semantics + crate API reference979db9b
26. firmware: harden buffered clip playback (adversarial review)e8909fd
27. firmware: clip reference client + protocol doc752fc28
28. firmware: clip playback engine + command dispatch92fb50f
29. firmware: CLIP protocol constants + status packerc99630c
30. firmware: clip SPSC byte ring (host-tested)39d2651
31. firmware: clip entry codec (host-tested)252df80
32. docs: buffered clip playback implementation plan3e29317
33. docs: buffered clip playback design spec17f8cd0
34. firmware: persistent, host-settable box name671afbe

Library

1. docs: fix README to the v3 unified-usage API0a34b65
2. fix: repair crate CI for v39eaed71
3. hw_full: correct stale proto assertion and grab all composite clone nodese724e36
4. python: clip trigger set + queryable config bindings59b2682
5. capi: clip trigger set + queryable config C ABI0663193
6. clip: trigger set, retain/rewind, and queryable config surface89c5521
7. cleanup: strip AI-slop from README and a stray doc-comment5f5b08e
8. cleanup: strip unnecessary comments and AI-slop (crate, capi, python)db05194
9. finish the builder unification + align the usage id fielddc0b940
10. bindings: rename MediusInput to MediusUsage (one Usage vocabulary)9c9bbd7
11. review: LockScope sum type, shared usage codec, safer conversions7657b2b
12. v3.0.0b2f684c
13. hw_full: v3 handshake + correct the clip/lock/name checks65ec4a7
14. host: clip held state is a usage snapshot across crate, C ABI, and Pythoncf2befe
15. python: unify bindings on the Input usage taxonomy (ABI v3)b8eaab3
16. capi: unify the C ABI on the Usage taxonomy (ABI v3)3659dee
17. crate: unify the input API on Usage and Axis (proto v3)0676994
18. clip: ClipConfig options type + field-generic control API6ac4187
19. clip: held is a per-button bitmask; enforce the edge cap; validate C edge actionsc47cfb1
20. crate + C ABI + Python: field-generic clip reuses existing typesa7158f3
21. docs: buffered clip playback section in the crate README5a5f16f
22. crate + C ABI + Python: buffered clip playback05728aa
23. crate + C ABI + Python: box name19c9f17

Docs

1. docs: update README to the v3 API (opcode range 0x15, all 5 sections, input-passthrough, drop dead MidnightUI link)7339e79
2. chore: shorten server.json description to the registry 100-char limit8474b06
3. chore: MCP registry manifest (server.json) + publishing guide3b74cc5
4. docs: demonstrate a move and an edge on one clip frame, all four surfacese9d7574
5. docs: move AI page to its own /ai section, add dropdown link12a3ef9
6. docs: fix clip reconnect claim and clip-query section polishbfc3014
7. docs: lay out the clip streaming/retained and trigger sections visually42973bd
8. docs: add the AI & LLM access page6ec7aea
9. feat: advertise MCP, cache headers, rel=alternate, prerender Homeb8c01e5
10. feat: ChatGPT-compatible search/fetch MCP tools6600245
11. docs: fix the AI-actions menu button font49f6ff3
12. fix: review findings (head defaults, search cap, route decode)275e7f4
13. feat: serve the agent surface under vite dev and previewfa14b49
14. ci: prerender stage in Docker, tests in CI, prerender scripts8044820
15. feat: move AI-actions to a titlebar dropdown (portaled)8fd103e
16. feat: per-route head + AI-actions row + SSG double-mount fixb4a4dc7
17. fix: harden markdown extractor (review findings)4b3e76d
18. feat: remote MCP docs server on /mcp2db15e8
19. feat: agent content-negotiation in the Bun serverc58e47f
20. feat: build-time prerender + markdown extraction pipeline and agent artifacts0e254db
21. docs: link the clip queries from the Clip page, like Lock and Catchcdd428d
22. docs: agent-accessible docs design spec3b67739
23. docs: move the clip queries onto the Requests page with the others205b7de
24. docs: full page-by-page consistency pass across all 78 pages79f41fc
25. docs: fix broken binding anchors + em-dash labels, mirror c/python clip API5294fa5
26. docs: tighten the clip page to the sibling norm82db2b0
27. docs: fix clip content + anchors from the migration review9ecaf50
28. docs: clip trigger set + queryable config across the SPA86a81be
29. docs: add the missing CLIP entry to feature-index enumerations950e079
30. cleanup: strip AI-slop and narrating comments from the docs-SPA components2a82ba0
31. docs: dashboard UI copy pass (terminology, labels, one diagram prop)9555b82
32. docs: page-by-page quality pass (dedup, links, diagram redraws)18fee2f
33. docs: finish the builder + field-rename propagation, ClipState meaning tables, reconnect badge265cdff
34. docs: apply deep-review findings + propagate the Usage rename52ab2c5
35. docs: v3.0.0 / proto 3 version references6db69c3
36. docs: unify the input taxonomy across the reference and dashboard39b49e5
37. docs: field-generic clip control and ClipConfig89ae248
38. docs: correct clip status held (button bitmask) and ticks (content-only)373f8f8
39. docs: rebuild the clip pages to match the rest of the docs, diagram-first831879d
40. docs: rewrite the clip pages, field-generic and tersebe75744
41. docs: buffered clip playbackea11667
42. docs + dashboard: box name1ed73a2

Firmware

1. firmware: soft-release clears a force-release, not just a press (v2.3.2)da24f52

Docs

1. docs: soft-release defers to physical from any state; version strings to 2.3.25e2d10a

Firmware

1. firmware: catch and lock cover multi-interface mouse dongles (v2.3.1)8a3c63f

Docs

1. docs: update firmware version strings to 2.3.13d77801
2. docs: correct the stale crate version (2.0.0 -> 2.3.0)1886e43
3. docs: bare-symbol card titles for discovery + a few consistency fixes8c88a19

Firmware

1. firmware: self-heal a control decoder wedged by an abandoned partial frameabb0c95
2. docs: sync the control-protocol reference + READMEs to the DEVICE_INFO rename (kind, product, box MAC)64dd946
3. firmware: name the DEVICE_INFO product-tail cap (127) so host string buffers keep NUL roomdc36571
4. tools: multi-box discovery (list / open-by-id) + generic device_info in the reference clientbd283dd
5. firmware: VERSION carries the box MAC; MOUSE_INFO -> generic DEVICE_INFO (+kind, +product)691aadb
6. docs: plan multi-box + generic device-info80a07bd
7. docs: spec multi-box management + generic device-info406a80a
8. host: robust HID report-descriptor fetch (blind-fetch + retry, verbatim)1158d0a
9. host: emit the clone snapshot once all HID interfaces are capturedf2642eb
10. host: count HID interfaces in a config descriptor (for capture completion)35ba281
11. docs: plan the capture-robustness implementation065ba8d
12. docs: spec capture robustness for slow composite HID devices7c1237b
13. Bump version to v2.3.0307c21e
14. tools: harden the emit-rate test + validation rig998cd72
15. firmware: reset the interval emit-rate when the device detachesfaa0ee0
16. tools: emit-rate HW validation rig02da0af
17. tools: emit-rate option in the reference client8c3ed75
18. firmware: wire the emit-rate option command + queryfbda1e0
19. firmware: emit-rate pacing override (state + gate)7351086
20. firmware: add the emit-rate option to the control protocol7fcdb74
21. docs: plan the emit-rate option implementationb5151ce
22. docs: spec a selectable emit-rate optiond347419

Library

1. style: cargo fmt345ad1e
2. crate: discover example (enumerate boxes + open by kind/id); gate open_transport to the mock buildabcfe0e
3. capi + python: device_info (+kind +product) + multi-box discovery (list/open_by_id/find_*_box) + version MACf642a95
4. crate: multi-box discovery (list/open_by_id/find_*_box) + reconnect by identity; device_info replaces mouse_info8480595
5. Bump version to 2.3.0d2f3a86
6. mock: match the firmware's emit-rate snapping1eb5851
7. emit-rate pacing: C ABI + Python bindings6d9ab05
8. emit-rate pacing API93b19fc
9. bindings(python): add a README and use it as the PyPI long descriptiond304bea

Docs

1. docs: update example version strings from 2.2.1 to 2.3.09eeb739
2. docs: bump docs site to v2.3.097a25f3
3. dashboard: flush a wedged box decoder before retrying the handshake7529503
4. dashboard: capitalize the device-kind chip, make the emit-rate radio vertical to match the others, and group Box id with the identity rows95124f5
5. docs: add the mac field to mock Version examples; correct is_composite to n_hid>1c61e317
6. document multi-box discovery + generic device-info (kind, product, box MAC)72f81a7
7. docs: rewrite the native Options page table-firstf78e1e5
8. document the emit-rate option218cf4e
9. dashboard: add the emit-rate option control39146b2

Firmware

1. firmware: binary hardening (v2.2.1)da48813
2. release: drop the parenthetical from the Library changelog header50d2233
3. release: combine the 3-repo changelog and attach a bin zip to Discord9157dd8

Library

1. bindings(python): add a README and use it as the PyPI long descriptiond304bea
2. ci: drop sccache (measured ~no gain) + the redundant host rust reinstall47ab1dc
3. ci: cache the wheel Rust builds with sccache (GHA backend)e1ad8b3
4. ci: aggressive caching to cut CI time32fa151
5. ci: attach only wheels to the release, not the sdistdc1fdde
6. ci: export MACOSX_DEPLOYMENT_TARGET=11.0 to cibuildwheel so delocate accepts the macOS wheels772f1ce
7. ci: pin macOS deployment target so the x86_64 wheel passes delocate6de7f4c
8. ci: wheels match the C tarball platforms (drop i686, add macOS x86_64)5bab67e
9. ci: skip-existing on PyPI publish so a re-dispatch is idempotentff4fa90
10. ci: fix the crates.io version check (User-Agent) + attach wheels to the release8940488
11. ci: build all 5 C ABI tarballs + add a manual backfill pathcc0dcac
12. ci: merge the C ABI + Python checks into ci.yml; gate the release on themfc94810
13. ci: drop the unavailable macos-13 wheel runner146fc21
14. release: bump the crate, C ABI, and Python binding to 2.2.117f04ba
15. bindings: drop the C++ wrapper; the C header serves C and C++336b4d4
16. test: cut drift — src/tests covers only what hw_full can't reachab361dd
17. Drop the docs/ directory — docs live in the medius-docs repo1c3d127
18. review: cut slop and dedupe across the bindings1d2cd64
19. docs(bindings): drop DIFFERENCES.md for a short README section; trim prose079731d
20. bindings: drop the Conan recipe / vcpkg plan04e8cf8
21. ci: fold the bindings release into the crate publish flow4ff3da7
22. ci(wheels): install libudev in the manylinux container; skip musllinux/win325b6c99f
23. bindings: deviation doc + publishing (PyPI, C/C++ release assets, Conan)eedbae9
24. bindings: restore the Clone capability (Device/streams/MockBox)c78696e
25. docs: bindings README + main README pointer to the C/C++/Python bindingsd044b5f
26. ci: build/test/lint the C ABI + bindings; fix the publish version check6a21cf9
27. bindings(python): self-contained sdist, py3-none wheel, API fixes20beb94
28. bindings(python): ctypes package over the C ABI + wheel packagingd3a652e
29. bindings(cpp): header-only RAII C++ wrapper over the C ABI00b5c36
30. capi: cap catch-event key count at u8::MAX so it can never wrap95f8113
31. capi: C ABI crate (medius-capi) covering the full medius surface2eeadb8
32. docs: design spec for the C ABI + C++/Python bindingscb6919b

Docs

1. docs: de-AI the binding prose and set version strings to 2.2.1478e9ec
2. docs: syntax-highlight code blocks with Prisme009fb4
3. docs: bindings section for C/C++ and Python49fede4
4. dashboard: render the grouped per-repo changelog headers9123bc2
5. docs: add a Discord link (top-bar button + search)52cfbc9