Fix Codex limits, Claude counting, and agent usage reliability from community PRs by dhh · Pull Request #14049 · omacom/omarchy

dhh · 2026-10-02T12:06:34Z

Brings in 25 community PRs that each fix a confirmed problem in the agent usage collectors or the agents panel. Each is one commit under its author (squashed where it had fixups), adapted where current quattro needed it, with the adaptation described under "Adapted:" in that commit. Every commit passes the agent tests on its own.

Codex

Fix Codex limits timing out on buffered app-server replies #13703 (tossbaws): read app-server replies from the raw fd, so batched or interleaved replies no longer time out the limits probe.
Resolve Codex through mise which instead of running the lazy launcher #13109 (surim0n): find Codex with mise which instead of running the lazy launcher, which installed Codex on refresh.
Skip the Codex app-server probe when there are no credentials #13106 (surim0n): skip the probe when the home being probed has no credentials. Adapted to check each account's own home, report "Waiting for auth", and read cli_auth_credentials_store with tomllib.
Fix Codex limits collector for CLI approval-policy churn #8977 (anonwurcod): detect a dead app-server and show the CLI's own error instead of a bare method name; quattro's approval flag is kept.
fix(agents): count pi sessions when HOME is a git checkout #13209 (Kevin Rajan): count pi sessions when $HOME is a git checkout.
Count only subscription-backed sessions as Codex usage #12032 (Nate Ashby): count only OpenAI-backed native sessions as Codex usage.
Deduplicate Codex usage across Pi forked sessions #8602 (Aris Gysel): dedupe pi usage across forked sessions.
Skip unchanged native Codex token snapshots #10531 (Brams): skip unchanged native token snapshots.
Count omp and pi profile sessions in the agent usage collectors #9546 (This_Is_NPC): count pi and omp profile sessions (Claude and Codex).
Pre-filter Codex session lines before JSON parsing #12803 (sanjyay): skip unrelated session lines before JSON parsing. Adapted to let session_meta through.
Read only the session files that changed since the last scan #12595 (PapistProtocol): read only the native session files that changed since the last scan.

Claude

Count streamed Claude messages by their highest-output usage line #10606 (steez): count streamed messages by their highest-output usage line, fixing undercounted output tokens.
Index Claude transcripts so the agents refresh reads only what was appended #8313 (Ryan Yogan): index transcripts so a refresh reads only what was appended. Adapted to keep the highest-output rule, re-read everything when the timezone changes, and rescan a transcript that was replaced rather than appended to.
Label a Claude Team seat by its subscription, not its rate-limit tier #11109 (Oli Denton) and Label the Claude plan from the profile the CLI refreshes #7225 (Igor Kramar): label Team seats by subscription and read the tier from the profile the CLI refreshes. Adapted so each account reads its own profile.
Call a lapsed Claude access token paused, not signed out #8093 (Martin Eidensten): an expired access token with a valid refresh token reads as paused, not signed out.
Refresh Claude usage after the clock moves backwards #9956 (Romain Perron): treat caches written in the future (clock moved back) as stale.
Bound unreadable Claude transcript warnings #12414: one bounded warning for unreadable transcripts.
Count Claude usage from opencode v2 sessions #13894 (manuaudio): count opencode v2 session_message rows.

Grok

A lapsed Grok access token with a refresh token is no longer treated as a sign-out. The token lives six hours and Grok renews it when it starts, so the panel was sending people through grok login several times a day. While Grok sits idle nothing on the machine spends its allowance, so the last limits show as current, and a weekly window that reset in the meantime starts over at 0%. A refresh token past Grok's 30-day sign-in asks for a sign-in again, and with nothing cached the section says to start Grok.

Panel

Add fallback reload Timer for agent usage records #10067 (fresh3nough): reload records that inotify missed; adapted to reload when each update run exits instead of on a timer, with a test that runs the real QML functions.
Stop the agents status card rendering as an empty box #8497 (Tyler South): show the status when the trouble line has no help text.
Clear stale agent login guidance after successful probes #8892 (whathek): clear the default login hint after a successful probe, extended to Grok.

Fireworks

Read Fireworks credentials from pi's auth.json #7455 (Ty Richards): read the Fireworks key from pi's auth.json.

Simplified

After the PRs and review fixes, a pass trimmed the collectors: the Claude index checks a digest of both ends of what it read alongside the file's identity, the Codex scan takes its reuse age as an argument instead of a global, unreachable guards in launcher detection are gone, and the scanner tests share one fixture setup.

Syncing usage between machines was left out: the headless snapshot publisher (#9525) and the guard on synced snapshot merging (#13339) are more machinery than the feature is worth.

Tests

All agent test files pass at every commit, plus new ones from the PRs and for the adaptations. ./test/cli passes apart from vscode generated theme references current theme file, which fails the same way on quattro. Reviewed with codex review --base quattro twice: the first pass's three findings (Claude index timezone and replaced files, single-quoted keyring setting) are fixed, and the second found nothing. Greptile's comments are fixed or answered. The panel changes (#10067, #8497) weren't checked in the running UI.

🤖 Generated with Claude Code

…#13109)

Adapted: credentials are checked in the home being probed rather than in the CODEX_HOME environment variable, since each registered account is probed in its own home, so a signed-out secondary account isn't hidden behind the primary's login. A home without credentials reports "Waiting for auth" like any other signed-out home. The credentials store setting is read with tomllib, so a single-quoted value counts too. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Detect an app-server that exits or stops answering, and report the end of its stderr instead of a bare RPC method name. Rebased onto the raw-fd reply reader; the switch from "-a on-request" to "-a never" is left out, keeping the current approval flags.

`omp --profile=<name>` (and pi's equivalent) relocates the whole agent tree under <base>/profiles/<name>/. The Claude and Codex collectors only ever scanned <base>/agent/sessions, so a subscription driven entirely through a profile was invisible to the agents panel: no tokens by day, no tokens by model, no prompt or session counts. Discover the profile roots alongside the default one. Sessions are keyed by file path, so a profile adds sessions instead of double-counting the default root, and a missing or unreadable profiles directory leaves the existing behavior untouched. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014zFbJcDEEpV5BAmsH6kAB3

Adapted: session_meta lines also pass the pre-filter, since the provider filter from #12032 reads them to skip rollouts served by a non-OpenAI provider. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

…12595) Native Codex rollouts keep per-file totals between runs, replayed while a file's mtime and size are unchanged. Rebased onto the session_meta provider filter, snapshot dedup, and line pre-filter, which now live in the per-file reader. pi and omp sessions are left out of the per-file cache: a forked pi session repeats its parent's messages, so they are deduplicated across the whole tree on every scan.

…0606) Claude Code writes a streamed assistant response as several transcript lines that share one message id, one per content block. Each line carries a usage object. The first line's output_tokens is a placeholder, often 1, and the last line has the real count. Input and cache fields usually match across the lines. The scanner dedupes by message id and keeps the first line it sees, so it under-counts output tokens. On a machine with 2,577 transcripts it reported 39.0M output tokens against 60.1M used, a 35% shortfall. Input and both cache fields differed by under 0.01%. Keep the line with the highest output count, with the last one scanned winning a tie. The whole line is kept because a response can fall back to another model mid-stream. Those lines are separate snapshots with different cache figures and a different model, and taking a maximum per field across them over-counts cache tokens and credits the wrong model. The zero-usage check now runs before dedup, so a zero-usage first line no longer claims a message id and hides a later line with real usage. Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> Co-authored-by: GPT-6 Astra <noreply@openai.com>

…pended (#8313) omarchy-agent-usage-claude re-parsed every line of every transcript under ~/.claude/projects on each refresh: no mtime cutoff, no memory of the last pass. The agents widget is on by default and ticks every 15 minutes, so the cost grew for the life of the machine. After one month here that was 803 files, 640 MB, 127k lines and 57k JSON parses per tick, about 1 core-second, pushed through the page cache every quarter hour forever. Keep a per-file index next to the scan cache: the unique usage records already parsed out of each transcript and the byte offset they end at. A file whose size and mtime match is not opened; a file that grew is read from the stored offset; a file that shrank or was rewritten is read from the start. --force drops the index and rescans from scratch. The summary is built from the indexed records in the same directory order the walk always used. That matters: when a resumed session carries earlier messages, the same message id appears in two files with different usage, and the first file visited wins. 91 ids differed on this machine; sorting the walk moved one model's output total by 25k tokens. Output is now byte-identical to the previous scan on a frozen copy of the corpus, cold, warm, and after an append. Warm refresh: 1.0 s -> 0.10 s of CPU, of which the scan itself is 70 ms; the index for this corpus is 2.9 MB. Adapted: - Rebased onto #10606: the highest-output rule for streamed messages now lives where the index parses records, and decides between files too. - The index records the timezone it was written in, and a change rereads every transcript, since its records hold local days. - A file only counts as appended to when its inode and the hash of what was already read still match, so a transcript replaced by a larger one, or rewritten in place, is read from the start. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

…#11109) The collector built the plan label from the OAuth rateLimitTier first, so a Team premium seat, which runs on default_claude_max_5x, showed in the agents panel as "Max 5x". Lead with subscriptionType and keep the multiplier as its qualifier: Max still reads "Max 5x", a Team seat reads "Team 5x". Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Adapted: the profile is found the same way current_account_id() finds it, now shared as profile_path(): ~/.claude.json for the default home, the home's own .claude.json otherwise. The original fell back to ~/.claude.json for any home without CLAUDE_CONFIG_DIR set, so a secondary account read the primary's tier. The profile's tier also keeps the subscription in the label, so a Team seat stays "Team" (#11109). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

…ommunity PRs (omacom#14049) * Read Codex app-server replies from the raw fd (omacom#13703) * Resolve Codex through mise which instead of running the lazy launcher (omacom#13109) * Skip the Codex app-server probe when there are no credentials (omacom#13106) Adapted: credentials are checked in the home being probed rather than in the CODEX_HOME environment variable, since each registered account is probed in its own home, so a signed-out secondary account isn't hidden behind the primary's login. A home without credentials reports "Waiting for auth" like any other signed-out home. The credentials store setting is read with tomllib, so a single-quoted value counts too. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Show the Codex CLI's own error when its app-server dies (omacom#8977) Detect an app-server that exits or stops answering, and report the end of its stderr instead of a bare RPC method name. Rebased onto the raw-fd reply reader; the switch from "-a on-request" to "-a never" is left out, keeping the current approval flags. * Count pi sessions when HOME is a git checkout (omacom#13209) * Count only OpenAI-backed native sessions as Codex usage (omacom#12032) * Deduplicate Pi usage across forked sessions (omacom#8602) * Skip unchanged native Codex token snapshots (omacom#10531) * Count omp and pi profile sessions in the agent usage collectors (omacom#9546) `omp --profile=<name>` (and pi's equivalent) relocates the whole agent tree under <base>/profiles/<name>/. The Claude and Codex collectors only ever scanned <base>/agent/sessions, so a subscription driven entirely through a profile was invisible to the agents panel: no tokens by day, no tokens by model, no prompt or session counts. Discover the profile roots alongside the default one. Sessions are keyed by file path, so a profile adds sessions instead of double-counting the default root, and a missing or unreadable profiles directory leaves the existing behavior untouched. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014zFbJcDEEpV5BAmsH6kAB3 * Skip unrelated Codex session lines before JSON parsing (omacom#12803) Adapted: session_meta lines also pass the pre-filter, since the provider filter from omacom#12032 reads them to skip rollouts served by a non-OpenAI provider. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Read only the Codex session files that changed since the last scan (omacom#12595) Native Codex rollouts keep per-file totals between runs, replayed while a file's mtime and size are unchanged. Rebased onto the session_meta provider filter, snapshot dedup, and line pre-filter, which now live in the per-file reader. pi and omp sessions are left out of the per-file cache: a forked pi session repeats its parent's messages, so they are deduplicated across the whole tree on every scan. * Count streamed Claude messages by their highest-output usage line (omacom#10606) Claude Code writes a streamed assistant response as several transcript lines that share one message id, one per content block. Each line carries a usage object. The first line's output_tokens is a placeholder, often 1, and the last line has the real count. Input and cache fields usually match across the lines. The scanner dedupes by message id and keeps the first line it sees, so it under-counts output tokens. On a machine with 2,577 transcripts it reported 39.0M output tokens against 60.1M used, a 35% shortfall. Input and both cache fields differed by under 0.01%. Keep the line with the highest output count, with the last one scanned winning a tie. The whole line is kept because a response can fall back to another model mid-stream. Those lines are separate snapshots with different cache figures and a different model, and taking a maximum per field across them over-counts cache tokens and credits the wrong model. The zero-usage check now runs before dedup, so a zero-usage first line no longer claims a message id and hides a later line with real usage. Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> Co-authored-by: GPT-6 Astra <noreply@openai.com> * Index Claude transcripts so the agents refresh reads only what was appended (omacom#8313) omarchy-agent-usage-claude re-parsed every line of every transcript under ~/.claude/projects on each refresh: no mtime cutoff, no memory of the last pass. The agents widget is on by default and ticks every 15 minutes, so the cost grew for the life of the machine. After one month here that was 803 files, 640 MB, 127k lines and 57k JSON parses per tick, about 1 core-second, pushed through the page cache every quarter hour forever. Keep a per-file index next to the scan cache: the unique usage records already parsed out of each transcript and the byte offset they end at. A file whose size and mtime match is not opened; a file that grew is read from the stored offset; a file that shrank or was rewritten is read from the start. --force drops the index and rescans from scratch. The summary is built from the indexed records in the same directory order the walk always used. That matters: when a resumed session carries earlier messages, the same message id appears in two files with different usage, and the first file visited wins. 91 ids differed on this machine; sorting the walk moved one model's output total by 25k tokens. Output is now byte-identical to the previous scan on a frozen copy of the corpus, cold, warm, and after an append. Warm refresh: 1.0 s -> 0.10 s of CPU, of which the scan itself is 70 ms; the index for this corpus is 2.9 MB. Adapted: - Rebased onto omacom#10606: the highest-output rule for streamed messages now lives where the index parses records, and decides between files too. - The index records the timezone it was written in, and a change rereads every transcript, since its records hold local days. - A file only counts as appended to when its inode and the hash of what was already read still match, so a transcript replaced by a larger one, or rewritten in place, is read from the start. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Label a Claude Team seat by its subscription, not its rate-limit tier (omacom#11109) The collector built the plan label from the OAuth rateLimitTier first, so a Team premium seat, which runs on default_claude_max_5x, showed in the agents panel as "Max 5x". Lead with subscriptionType and keep the multiplier as its qualifier: Max still reads "Max 5x", a Team seat reads "Team 5x". Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * Label the Claude plan from the profile the CLI refreshes (omacom#7225) Adapted: the profile is found the same way current_account_id() finds it, now shared as profile_path(): ~/.claude.json for the default home, the home's own .claude.json otherwise. The original fell back to ~/.claude.json for any home without CLAUDE_CONFIG_DIR set, so a secondary account read the primary's tier. The profile's tier also keeps the subscription in the label, so a Team seat stays "Team" (omacom#11109). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Call a lapsed Claude access token paused, not signed out (omacom#8093) * Refresh Claude usage after the clock moves backwards (omacom#9956) * Bound unreadable Claude transcript warnings (omacom#12414) * Count Claude usage from opencode v2 sessions (omacom#13894) * Reload agent usage records when an inotify watch fails to rearm (omacom#10067) * Reload agent usage records after each update run instead of on a timer Rather than omacom#10067's two-minute timer per record, reload every record when the omarchy-agent-usage-update process exits, the moment its files can have been replaced. A reload that finds a file unchanged keeps its record, so the panel isn't stirred up by identical data. The grep test now runs the QML functions. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Show the agent status when the trouble line has no help text (omacom#8497) * Clear stale agent login guidance after a successful probe (omacom#8892) * Clear the Grok login hint after a successful probe omacom#8892 cleared the default login hint after a successful probe in the Claude and Codex collectors; Grok's collector had the same stale hint. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Read Fireworks credentials from pi's auth.json (omacom#7455) The Fireworks collector skipped pi, Omarchy's default agent, when walking its credential ladder, so a machine signed in to Fireworks only through pi (/login fireworks) never showed the tab. Insert the key pi stores in $PI_CODING_AGENT_DIR/auth.json (default ~/.pi/agent) between the firectl auth.ini and the opencode fallback. pi keys can be literals, $ENV_VAR/${ENV_VAR} references, or !command shell lookups. The collector resolves the first two; command lookups stay pi-only and are skipped rather than sent to the API verbatim. * Call a lapsed Grok access token paused, not signed out Grok's access token lives six hours and Grok mints a new one from its refresh token whenever it starts, so a lapsed one is routine. Reporting it as an expired sign-in made the panel offer Sign-in required several times a day, sending people through grok login for nothing. With a refresh token present it now reads as paused, like Claude's. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Keep showing Grok's last limits while it sits idle While Grok hasn't run, nothing on the machine has spent its allowance, so with a refresh token on hand the last numbers still stand: they show as current rather than dimmed under a status line. A weekly window that reset in the meantime starts over at 0%, a whole number of weeks on. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Ask for a Grok sign-in once its refresh token is past 30 days A refresh token older than Grok's 30-day sign-in can't renew anything, so the panel offers Sign-in required again instead of showing the last limits as current. With nothing cached yet it says to start Grok, rather than showing an empty section without a word. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Check both ends of what the Claude index read before resuming a transcript A transcript rewritten in place could grow and change only after its first kilobytes, and the index took it for an append. It now compares the last kilobytes before the resume point too. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Simplify the agent usage collectors - Codex: pass the forced-scan choice down instead of a module global, make the per-file reader's cache arguments required, shrink the cache record check, and drop guards for shapes that can't occur: an empty launcher path, realpath raising, mise itself being a lazy launcher, multi-line `mise which` output, and probing without a temp file for stderr. - Claude: decide an append by the digest of both ends of what was read alone; the inode and mtime checks it made redundant are gone. - Snapshot: the device id falls back to the hostname, which always exists. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Share fixture setup in the agent usage scanner tests Every fixture home lives under one scratch directory with a single cleanup trap, instead of a trap rewritten with a longer list for each new home, and the Codex test builds its signed-in homes with one helper. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Treat a replaced Claude transcript as new even when its ends match Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Probe Codex without its error text when there's no temporary space Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: tossbaws <17258053+tossbaws@users.noreply.github.com> Co-authored-by: surim0n <suritech@gmail.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: anonwurcod <anonwurcod@proton.me> Co-authored-by: Kevin Rajan <7121943+kvnloo@users.noreply.github.com> Co-authored-by: Nate Ashby <nate.ashby11@gmail.com> Co-authored-by: Aris Gysel <aris.gysel@me.com> Co-authored-by: Brams <76213579+Brams-s@users.noreply.github.com> Co-authored-by: This_Is_NPC <gabrielfollone27@gmail.com> Co-authored-by: sanjyay <102979855+sanjyay@users.noreply.github.com> Co-authored-by: PapistProtocol <12738904+PapistProtocol@users.noreply.github.com> Co-authored-by: steez <stevedimakos97@gmail.com> Co-authored-by: GPT-6 Astra <noreply@openai.com> Co-authored-by: Ryan Yogan <ryanyogan@gmail.com> Co-authored-by: Oli Denton <41393837+omdenton@users.noreply.github.com> Co-authored-by: Igor Kramar <i@ikramar.ru> Co-authored-by: Martin Eidensten <martin@meibe.se> Co-authored-by: Romain Perron <rdj.perron@gmail.com> Co-authored-by: Omarchy Contributor <contributor@users.noreply.github.com> Co-authored-by: manuaudio <manu@arimaka.com> Co-authored-by: Tyler South <tsouth2@gmail.com> Co-authored-by: whathek <Hek846@users.noreply.github.com> Co-authored-by: Ty Richards <me@tyrichards.com>

tossbaws and others added 19 commits October 1, 2026 21:51

Read Codex app-server replies from the raw fd (#13703)

ab10a52

Resolve Codex through mise which instead of running the lazy launcher (…

8f4fcfa

…#13109)

Count pi sessions when HOME is a git checkout (#13209)

719ae7a

Count only OpenAI-backed native sessions as Codex usage (#12032)

26c5750

Deduplicate Pi usage across forked sessions (#8602)

feed004

Skip unchanged native Codex token snapshots (#10531)

5207546

Skip unrelated Codex session lines before JSON parsing (#12803)

02165e6

Adapted: session_meta lines also pass the pre-filter, since the provider filter from #12032 reads them to skip rollouts served by a non-OpenAI provider. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Call a lapsed Claude access token paused, not signed out (#8093)

3d01e99

Refresh Claude usage after the clock moves backwards (#9956)

b46980b

Bound unreadable Claude transcript warnings (#12414)

d38c79e

Count Claude usage from opencode v2 sessions (#13894)

0c60b94

dhh deleted the agent-usage-fixes branch October 3, 2026 02:03

This was referenced Oct 3, 2026

agents: Claude limits freeze permanently after a backward clock step (negative cache age reads as fresh) #12305

Closed

Agents panel reports SIGN-IN EXPIRED when only the 8-hour access token lapsed #12470

Open

omarchybot mentioned this pull request Oct 3, 2026

Read Claude limits with a sibling CLI's token when the saved one lapsed #8326

Open

cursor Bot mentioned this pull request Oct 3, 2026

Merge upstream quattro (Codex/Claude usage, OpenClaw self-update, image picker) mark-groves/omarchy#59

Merged

aholbreich mentioned this pull request Oct 3, 2026

Agents panel: "Codex limits unavailable" because account/read hangs; limits are fetchable #14120

Closed

cursor Bot mentioned this pull request Oct 4, 2026

Merge upstream quattro (fingerprint recovery, Microsoft web apps, leftover usage/OpenClaw/picker) mark-groves/omarchy#60

Merged

aron-intframe mentioned this pull request Oct 5, 2026

Codex collector: intermittent account/read timeout blanks meters and never retries #14293

Open

Navigation Menu

Repository navigation

Fix Codex limits, Claude counting, and agent usage reliability from community PRs - #14049

Fix Codex limits, Claude counting, and agent usage reliability from community PRs#14049
dhh merged 33 commits into
quattrofrom
agent-usage-fixes

dhh commented Oct 2, 2026 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

Conversation

dhh commented Oct 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codex

Claude

Grok

Panel

Fireworks

Simplified

Tests

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

dhh commented Oct 2, 2026 •

edited

Loading