Conversation
…igration * ts(slice-16, wip 1/N): port peer-call/gemini (.sh→.ts) Sibling of slice-15 grok port (PR #896). Wraps 'gemini -p' non-interactive headless mode. Read-only safety preserved (--approval-mode plan + --skip-trust per the Copilot fix on the bash original PR #28). Reuses the structural pattern from grok.ts: - classifyFlag + MutableArgState parser (under cog-complexity 15) - spawnSync + stdio:inherit for live LLM streaming - isRegularFile + readHead helpers - /bin/sh -c for context-cmd (same trust as bash eval) - PREAMBLE preserved verbatim with Gemini-specific framing (proposes role per four-ferry consensus) Lint-clean: bun --bun tsc --noEmit + eslint strictTypeChecked + sonarjs all pass. Argument-validation byte-equivalent. Note: the same CodeQL js/indirect-command-line-injection alert will fire here as on grok.ts; per B-0107 (filed in PR #897) the per-PR dismissal pattern applies. Will dismiss after PR opens.
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 1bfc147e07
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
…through-head context-cmd Two real Codex findings on PR #898 (both also apply to grok.ts — backport in follow-up PR): P1 — Read only file head instead of loading full file: Original used readFileSync + Buffer.subarray which loads the entire file into memory then slices. For large artifacts (logs, dumps) this is a memory-spike regression vs bash 'head -c 20000'. Replaced with openSync + readSync(fd, buf, 0, bytes, 0) which reads only the first FILE_HEAD_BYTES bytes from disk. Wrapped in try/finally for fd cleanup. P2 — Truncate context command output at source: Original used spawnSync with maxBuffer up to 64MB then sliced to CTX_HEAD_BYTES afterward. For high-volume commands (wide 'git diff' ranges) this blocks longer and uses more memory than the bash original's '... | head -c 20000' pipeline that short-circuits at the boundary. Reframed wrapped command as: (cmd) 2>&1 | head -c <N> so the shell pipeline truncates at source. Same trust contract preserved (user supplies cmd; we just augment with head-c). Verified locally: 100MB test file reads only 20000 bytes via the new head-only path.
There was a problem hiding this comment.
Pull request overview
Ports the peer-call Gemini proposer script from bash to TypeScript/Bun as part of the ongoing TS/Bun migration, and updates the migration trajectory status to reflect the new slice in flight.
Changes:
- Add
tools/peer-call/gemini.tsimplementing the Gemini peer-caller in TS/Bun (arg parsing, prompt construction, optional file/context attachments,gemini -pinvocation). - Update the TS/Bun migration RESUME trajectory status/milestone for slice 16.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 5 comments.
| File | Description |
|---|---|
| tools/peer-call/gemini.ts | New TS/Bun Gemini peer-call script (port of gemini.sh) with streaming CLI invocation and prompt assembly. |
| docs/trajectories/typescript-bun-migration/RESUME.md | Updates trajectory status/milestone to include slice 15 merged and slice 16 in-flight. |
4 tasks
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: be2507fd6f
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
Both reviewers caught: runContextCmd was returning only result.stdout, which drops diagnostics emitted on the shell's OWN stderr (parse/syntax errors in --context-cmd). The (cmd) 2>&1 redirect only captures the command's stdout/stderr; the SHELL's stderr (where parse errors land) falls outside that redirect. Fix: concat result.stdout + result.stderr, then slice to CTX_HEAD_BYTES. On a clean run stderr is empty so output is unchanged; on malformed cmd the parse error reaches the prompt instead of an empty context block. Same fix needed on gemini.ts in #898 — opening a follow-up commit there.
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
…read + pipe-through-head context-cmd (#899) * fix(peer-call/grok): backport Codex P1+P2 from #898 — head-only file read + pipe-through-head context-cmd Backports the same two perf fixes that landed on gemini.ts in PR #898 — the bugs are identical because grok.ts and gemini.ts share the same readHead + runContextCmd helper shape (sibling-port pattern from slice-15 / slice-16). P1 — Read only file head instead of loading full file: Original used readFileSync + Buffer.subarray which loads the entire file into memory then slices. For large artifacts (logs, dumps) this is a memory-spike regression vs bash 'head -c 20000'. Replaced with openSync + readSync(fd, buf, 0, bytes, 0) which reads only the first FILE_HEAD_BYTES bytes from disk. Wrapped in try/finally for fd cleanup. P2 — Truncate context command output at source: Original used spawnSync with maxBuffer up to 64MB then sliced to CTX_HEAD_BYTES afterward. For high-volume commands (wide 'git diff' ranges) this blocks longer and uses more memory than the bash original's '... | head -c 20000' pipeline that short-circuits at the boundary. Reframed wrapped command as: (cmd) 2>&1 | head -c <N> so the shell pipeline truncates at source. Same trust contract preserved (user supplies cmd; we just augment with head-c). Same fixes verified on gemini.ts in #898 (100MB test file reads only 20000 bytes via the new head-only path). * review(#899): preserve shell parse errors per Codex P2 + Copilot Both reviewers caught: runContextCmd was returning only result.stdout, which drops diagnostics emitted on the shell's OWN stderr (parse/syntax errors in --context-cmd). The (cmd) 2>&1 redirect only captures the command's stdout/stderr; the SHELL's stderr (where parse errors land) falls outside that redirect. Fix: concat result.stdout + result.stderr, then slice to CTX_HEAD_BYTES. On a clean run stderr is empty so output is unchanged; on malformed cmd the parse error reaches the prompt instead of an empty context block. Same fix needed on gemini.ts in #898 — opening a follow-up commit there.
…on, bash shell, file-read errors Five real findings from Copilot+Codex on PR #898: Copilot P1 — Exit codes 0|2|64 per repo-scripting.md: Switched return 1 → return 64 for invocation/usage errors (--model/--file/--context-cmd missing values, unknown flag, empty prompt). Aligns with the conventions in docs/best- practices/repo-scripting.md §exit-codes. Process-related errors (gemini missing on PATH, build-prompt error) keep 1 for tooling/input failure. Copilot P1 — spawnSync result.status ?? 1 collapses launch failures: Added classifySpawnFailure helper (4-case: status set / ENOENT → 127 / signal / other) — same pattern from PR #887. Distinguishes ENOENT/permission/signal from a normal non-zero exit, with a contextual stderr message. Codex P2 — eval vs /bin/sh syntax difference: Switched /bin/sh -c → /bin/bash -c so the bash original's eval semantics (accepting bash-only features like `[[ ]]`, brace expansion, process substitution) are preserved. /bin/sh on Ubuntu is dash and rejects these. Codex P2 — file-read failures silently dropping context: readHead() now returns ReadHeadResult { ok, content, error }. buildFullPrompt() propagates the error to the user via stderr instead of pretending context was attached when read failed (permission denied, race, etc.). Threads on already-addressed findings (PRRT_kwDOSF9kNM5-pd16 runContextCmd full-buffering + PRRT_kwDOSF9kNM5-pd2a readHead full-file-read) were resolved in the parse-error-fix commit ceed1e7. Threads on PRRT_kwDOSF9kNM5-pd2U PREAMBLE-names: pushing back because the PREAMBLE is prompt content sent verbatim to the LLM peer, not pure code-surface; changing it diverges from the bash original byte-equivalence. A separate task to retool PREAMBLE attribution on both bash AND TS together is the right shape, not a one-side rename here.
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: c46d1792f0
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Three new findings on the latest push:
Codex P2 (PRRT_kwDOSF9kNM5-pon2) + Copilot P0 x2 — exit codes:
Last commit switched to exit 64 per docs/best-practices/repo-
scripting.md. Reviewers correctly point out tools/peer-call/
README.md is the more specific spec and says 0/1/2 uniform
across all three peer-call wrappers. Specific wins on overlap.
Reverted ArgError.exitCode 64 → 1 + the empty-prompt return
64 → 1. Matches grok.ts (slice 15) and the bash original.
Copilot P1 — commandAvailable() shape:
Was using `<cmd> --version` and requiring exit 0. Some CLIs
exit non-zero on --version. Bash uses `command -v <cmd>`
(PATH existence check, no execution). Switched to
`spawnSync('/bin/sh', ['-c', \`command -v "${cmd}"\`])`
to match bash semantics.
Copilot P1 (PRRT_kwDOSF9kNM5-ppuR) — PREAMBLE names: same as
prior thread; pushing back with same rationale (PREAMBLE is
verbatim prompt content sent to LLM, not pure code-surface;
preserving bash-equivalence). Will reply on the new thread
the same way.
Copilot P2 (PRRT_kwDOSF9kNM5-ppt6) — PR description says /bin/sh
but code uses /bin/bash. The /bin/bash switch was a Codex P2 from
the prior round (preserve eval semantics for bash-only features).
Will update PR description, not the code.
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
4 tasks
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
…igration (#900) * ts(slice-17, wip 1/N): port peer-call/codex (.sh→.ts) Third sibling of the peer-call cluster (after grok #896 and gemini #898). Wraps 'codex exec -s read-only --skip-git-repo-check' (default) or 'codex review' (with --review). --model is ignored in --review mode with a warning to stderr. Bakes in all the round-2 + round-3 fixes from the gemini.ts review cycle: - exit codes 0/1/2 uniform per tools/peer-call/README.md (NOT 64) - commandAvailable via /bin/sh -c 'command -v <cmd>' (PATH check, not --version) - /bin/bash -c (not /bin/sh -c) for context-cmd to preserve bash-only feature support (Codex P2) - ReadHeadResult { ok, content, error } so file-read failures surface to stderr instead of silently dropping context - classifySpawnFailure 4-case helper (status / ENOENT → 127 / signal / other) reused from PR #887 + slice-16 #898 - Concat stdout + stderr from runContextCmd to preserve shell parse errors that fall outside the inner ( ) 2>&1 redirect Lint-clean: bun --bun tsc --noEmit + eslint strictTypeChecked + sonarjs all pass. Argument-validation byte-equivalent. Note: same CodeQL js/indirect-command-line-injection alert will fire here as on grok.ts and gemini.ts. Per B-0107 (filed in PR #897) the per-PR dismissal pattern applies until B-0107 implements the structural fix.
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
…ull-stream guards + header phrasing Three Copilot findings on #901: P0 — spawnSync launch failures collapsed into exitCode 1: Added classifySpawnFailure helper (4-case: status set / ENOENT → 127 / signal / other) reused from PRs #887, #898, #900. Both runSnapshotBurn and runProjectRunway now report a contextual note when the child can't start (e.g., 'snapshot-burn.sh: command not found on PATH (ENOENT)'). P0 — null stdout/stderr could yield 'nullnull': When a child fails to start, result.stdout / result.stderr can be null even with encoding: 'utf8'. Guarded with `?? ''` in runProjectRunway so the projection block doesn't end up as the literal string 'nullnull'. P2 — Header comment phrasing: Reworded 'snapshot-burn.sh ported in #894' to 'TS port snapshot-burn.ts landed in #894 but this wrapper still spawns the .sh during the soak period' to avoid implying the .sh file itself is the ported artifact.
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
…ary-condition - Extracted runSnapshotStep + runProjectionStep helpers to drop main() under cognitive-complexity 15. - Added eslint-disable on stdout/stderr ?? '' guards (typings claim string when encoding is set, but the runtime can return null when spawn fails — same pattern as #898). Lint clean: tsc + eslint strictTypeChecked + sonarjs all pass.
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
…tion (#901) * ts(B-0086): port 1 budget script (.sh→.ts) — slice 18 of TS/Bun migration * ts(slice-18, wip 1/N): port budget/daily-cost-report (.sh→.ts) Daily cost-monitoring entry point. Wraps snapshot-burn.sh + project-runway.sh and writes human-readable projection to docs/budget-history/latest-report.md (visibility surface for Aaron's #287 deadline). Note: this wrapper still spawns the bash siblings (snapshot-burn.sh + project-runway.sh), NOT the TS port — the bash versions are the soak-period reference until they retire. Once project-runway is also TS-ported, this wrapper can switch to spawning the .ts versions. Mechanical changes: - bash arg-parse → parseArgs with ParsedArgs | ArgError | help - bash 'cat > "$report_path" <<EOF...EOF' heredoc → buildReport() template literal returning the markdown string - bash subshell command capture ($(...)) → spawnSync with stdio ['inherit', 'pipe', 'pipe'] for project-runway combined output - bash heredoc concat across multi-line → resolved via separate argsSuffix variable (sonarjs no-nested-template-literals) - exit codes 0/1/2 preserved verbatim per bash original Lint-clean: tsc --noEmit + eslint strictTypeChecked + sonarjs all pass. Argument-validation byte-equivalent. Trajectory: 39 ports total after merge, 4 Bucket B files remain (1 budget project-runway + 1 git/batch-resolve + 1 pr-preservation + 1 in-flight = 4 remaining). * review(slice-18): use statSync.isFile() instead of existsSync per Codex P2 Codex P2: existsSync returns true for directories and other non-regular paths; the bash original uses -f which checks regular-file existence. If snapshots.jsonl were ever a directory, existsSync would skip the bootstrap branch and the wrapper would try to spawn project-runway.sh against a non-file. Switched to statSync.isFile() with try/catch fallback to false. * review(slice-18): address Copilot P0+P0+P2 — spawn classification + null-stream guards + header phrasing Three Copilot findings on #901: P0 — spawnSync launch failures collapsed into exitCode 1: Added classifySpawnFailure helper (4-case: status set / ENOENT → 127 / signal / other) reused from PRs #887, #898, #900. Both runSnapshotBurn and runProjectRunway now report a contextual note when the child can't start (e.g., 'snapshot-burn.sh: command not found on PATH (ENOENT)'). P0 — null stdout/stderr could yield 'nullnull': When a child fails to start, result.stdout / result.stderr can be null even with encoding: 'utf8'. Guarded with `?? ''` in runProjectRunway so the projection block doesn't end up as the literal string 'nullnull'. P2 — Header comment phrasing: Reworded 'snapshot-burn.sh ported in #894' to 'TS port snapshot-burn.ts landed in #894 but this wrapper still spawns the .sh during the soak period' to avoid implying the .sh file itself is the ported artifact. * review(slice-18) round-2: extract step-helpers + suppress no-unnecessary-condition - Extracted runSnapshotStep + runProjectionStep helpers to drop main() under cognitive-complexity 15. - Added eslint-disable on stdout/stderr ?? '' guards (typings claim string when encoding is set, but the runtime can return null when spawn fails — same pattern as #898). Lint clean: tsc + eslint strictTypeChecked + sonarjs all pass. * review(slice-18): preserve stdout/stderr ordering via shell-side 2>&1 per Copilot P1 Copilot caught: concatenating result.stdout + result.stderr does NOT preserve the original chronological ordering of merged streams. The bash original $(... 2>&1) merges at the kernel pipe level — if project-runway.sh emits warnings on stderr while writing success output to stdout, the messages interleave correctly. Switched to /bin/bash -c 'path 2>&1' so the merge happens shell-side (matches bash original semantics). Single stdout pipe = correct ordering. result.stderr is now unused (the inherit pipe still receives nothing).
4 tasks
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
…sure (#903) Per the consolidated-row-pattern (rows from 03:41Z + 05:01Z arc closures): when a session lands many small commits across multiple ticks, a single consolidated row summarizing the arc is more signal-dense than N minimal rows. Covers the slices 15-19 arc that landed after the 05:43Z slice-14 row: - #896 slice 15 (grok.ts) — peer-call cluster opens - #898 slice 16 (gemini.ts) — peer-call sibling - #899 backport to grok from #898 review-cycle findings - #900 slice 17 (codex.ts) — peer-call cluster closes - #897 B-0107 row — CodeQL dismissal pattern - #901 slice 18 (daily-cost-report.ts) — budget wrapper - #902 slice 19 (project-runway.ts) — budget cluster closes (in flight) Three new substrate observations recorded for future-Otto: - sibling-port-cost decreases monotonically (round-2/3 fixes bake into later siblings proactively) - kernel-pipe vs JS-space stream interleaving distinction (bash `2>&1` merges shell-side; `result.stdout + result.stderr` in JS does not preserve chronological ordering) - fix-the-bug + file-the-row + implement-the-row + closeout pattern is the durable shape (B-0106 + B-0107 are both worked examples) Cron 98fc7424 still armed. Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
2 tasks
AceHack
added a commit
that referenced
this pull request
Apr 30, 2026
…s updates (#904) * docs(ts-bun-migration): backfill slice-16/17/18 audit entries + status updates Backfills the slice-audits.md gap surfaced during the slice-19 close: slices 16, 17, 18 had merged but their audit entries were missing. Also flips slice 15 + slice 19 entries from "PR pending" to merged status (PR # + commit SHA). Slice 16 (peer-call/gemini, #898): three rounds of review-cycle fixes documented (round 1 exit-code 1 + commandAvailable PATH check; round 2 /bin/bash -c + shell-side truncation + ReadHeadResult; round 3 stderr concat for parse-error preservation). Slice 17 (peer-call/codex, #900): closes peer-call cluster. Documents the sibling-port-cost-decreases-monotonically pattern — slice 17 shipped 357 lines in a single commit with all known fixes pre-baked. Slice 18 (budget/daily-cost-report, #901): first wrapper-class port. Documents the kernel-pipe vs JS-space stream-ordering distinction that became substrate observation in the 07:21Z tick row. No code changes. Doc-only. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(ts-bun-audits): address #904 review threads — MD038, persona attribution, monotonic deltas Round-1 fixes for PR #904: - MD038 markdownlint failure (line 451) — slice-18 entry had nested backticks (` "${path}" 2>&1 ` inside an outer code span) that confused markdown's parser. Rephrased prose to avoid embedding backticks inside backticks. - Persona-name attribution in current-state doc (Copilot P1, twice) — slice-16 + slice-17 file descriptions used "Otto's harness-side caller for invoking ..." Per the .github/copilot-instructions.md no-name-attribution rule for current-state docs, switched to role-ref "the harness-side caller for invoking ...". Slice-15 pre-existing wording on main left alone (separate commit if needed). - Non-monotonic Bucket B deltas (Codex P2) — backfilled rows for slices 16/17/18 had stale numbers (9→8, 8→7, 7→6) that were inconsistent with the pre-existing slice-15 row's "8 → 7". Recomputed to monotonic sequence: slice 16 = 7→6, slice 17 = 6→5, slice 18 = 5→4. - Slice-19 outcome wording (Copilot) — "Once this lands" → "Now that this has landed" since slice 19 already merged at commit bfdadd9. Doc-only. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Slice 16 — sibling port of grok.ts (slice 15 #896):
tools/peer-call/gemini.{sh→ts}(Gemini propose-role peer-caller via gemini CLI)Reuses the structural pattern from grok.ts:
classifyFlag+MutableArgStateparser,spawnSync+stdio: "inherit"for live LLM streaming,/bin/sh -cfor context-cmd. PREAMBLE preserved verbatim with Gemini-specific propose-role framing per the four-ferry consensus.Read-only safety preserved (
--approval-mode plan+--skip-trustper the Copilot fix on bash original PR #28).Test plan
bun --bun tsc --noEmitclean (lint-tsc-tools gate from ci(B-0106): add lint-tsc-tools gate job — tsc --noEmit on tsconfig.json #890 will validate this in CI).eslintclean.--modelwithout value → exit 1; no prompt → exit 1.Note on CodeQL alert
The same
js/indirect-command-line-injectionalert that fired on grok.ts (#896) will fire here onrunContextCmd. Per B-0107 (filed in #897) the per-PR dismissal pattern applies until B-0107 implements the structural fix.Trajectory
docs/trajectories/typescript-bun-migration/RESUME.md.🤖 Generated with Claude Code