4 Commits
Author SHA1 Message Date
Aleksander DygońandClaude Opus 5.5 03c2ff34c2 Remove code comments and add the no-comments rule (#12604)
* Remove code comments and add the no-comments rule

Remove the comments from first-party code and configuration files:
TS/JS, JSX, CSS, SQL migrations, Swift, Kotlin, Gradle, shell,
Docker, systemd, Nix, YAML, TOML, JSONC and GritQL.

Keep only what a tool reads: license headers, shebangs, compiler
and lint directives, JSDoc types in plain .js files, generated file
headers, GritQL scope lines, the install-server.sh help text and the
openbot-boat-pre-stop marker. Vendored code, fixtures and released
schemas keep their comments.

Add the rule to AGENTS.md, and change the docs that asked for
comments. Add two i18n keys that only comments named to
RELEASED_SOURCE_KEYS.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Keep comments that are data in strings

The benchmark task files in opencode-free-models-bench.ts tell the
model what each function must do. The sample file and HTML page in
the two stories are display content. Put their comments back.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Keep released migrations and mobile native inputs unchanged

check:migrations compares each released migration with v0.73.0 byte
for byte, so restore src/backend/migrations. A change to a native
build input of apps/mobile makes a new runtime version, so restore
those files and the extra fingerprint sources.

List these files and the comment markers that a tool reads in
AGENTS.md. Remove the last requests for comments in
src/renderer/AGENTS.md.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-19 12:19:28 +01:00
NB 7c3a9f4f6c Use less memory in tabs, provider processes and app windows (#5672)
* Use less memory in tabs, provider processes and app windows

- Keep only the 3 most recently used idle browser tabs loaded, and unload an idle tab after 15 min.
- Release idle Claude, ACP, Pi and Codex sessions after 5 min, and keep one idle session loaded.
- Stop the full process group of each provider CLI, so MCP server children do not stay.
- Load only English and the shown language in each app window, not all 15 languages.
- Keep fewer hidden side threads, chat images and long Markdown messages in the renderer.
- Send the JavaScript heap of the app windows in the daily resource analytics.

* Fix review findings in the memory changes

- Wait for the saved language before the first render.
- Do not stop ACP agents as a process group: some leave the user's servers in the agent's group.
- Close other idle Pi threads in the background, so a failed close does not fail the turn.
- Release idle threads once in each critical memory period, also when a release was in progress.
- Forget the idle Codex sessions of a stopped client.
- Share the lazy language catalog Vite plugin from packages/i18n.
- Limit the Markdown lexer cache by characters, not by body size.

* Do not export types that only their own file uses

* Fix CI: type ratchet and the Antigravity MCP load failure test

- Set `resumedSession` only when the session resumed, for exactOptionalPropertyTypes.
- Only one idle ACP thread keeps its process, so the first thread can resume once more in the
  last turn of the Antigravity test. The two failed-turn resumes are still checked exactly.
2026-11-08 00:54:40 +01:00
NBandClaude Opus 5.5 be48a4c1b8 Send a daily release regression report to Discord (#5104)
* Send a daily release regression report to Discord

A scheduled workflow compares the two newest releases from production
analytics: turn time, failed turns, event loop, database time and memory.
The table goes to a private Discord channel. The repository is public, so
the log shows no data and the report is not uploaded.

CI connects with its own SSH key as a user that can only forward to a
local proxy of the read-only ClickHouse user. The secrets are in the
`analytics` environment, which only protected branches can use.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Show the release regression report as colored Discord cards

Each line has a status: red for a regression, yellow for a smaller
worsening to watch, green for no change or better, and grey when the
sample cannot show it yet. A summary card counts the statuses and lists
the red and yellow lines first. Each section card has the color of its
worst line.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Add crashes, updates, error rates and startup to the release report

- Renderer crashes per computer-day, update failures, and error codes
  and turn failure causes per 1,000 turns.
- Startup and first output rows, when both releases send them.
- Shares and error rates need a significant rise (one-sided z test).
- Failed turns "OpenBot causes" leave out the outside causes of #5053.
- The scheduled run sends only when a status changes, and on Monday.
  The cache holds only an HMAC of the statuses.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Fix review findings in the release regression report

- Count an error code only when it has events on 3 computers or more.
- Show a waiting title when no metric is judged.
- Read the ClickHouse credentials once.
- Stop a Discord send after 30 seconds.
- Send nothing, and do not fail, when fewer than two releases have data.
- Keep descriptions within the Discord limit, and send no empty message.
- Post a Discord notice when the tunnel does not open.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Clear the release regression state after any unhandled failure

Restore the state before the tunnel opens. When the tunnel, a timeout or a
cancel stops the run before the script can clear the state, a workflow step
clears it and posts a notice. The report step gets its own time limit, so
that step has time to run.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 11:36:21 +01:00
NBandClaude Opus 5.5 807acc1149 Add read-only SQL access to the OpenPanel ClickHouse (#5046)
* Add read-only SQL access to the OpenPanel ClickHouse

The OpenPanel MCP has no percentiles and returns about 12 raw rows.
scripts/openpanel-sql.ts runs read-only SQL through an SSH tunnel as the
ClickHouse user openbot_readonly. That user can only read the openbot
rows of the events table. Presets give p50, p90 and p99 with sample sizes
for turn duration, system resources, conversation opens and voice turns.
Output redacts emails and profile, device and session IDs by default.

The credentials are encrypted in apps/auth-api/.env.dev and .env.production.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Show only the ClickHouse exception and keep identifier counts

- An error after partial output (result-row limit) showed the partial rows, with profile IDs, in the
  error message. Now the message has only the ClickHouse exception.
- Counts such as uniqExact(profile_id) stay numbers. Only text values of identifier columns are
  redacted.
- The docker inspect call over SSH has the tunnel deadline.
- --days must be 1 or more, and --days with --since or --until is an error.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 21:47:53 +01:00