docs(config): document RMP_ENGINE + RMP_DRIVER flags #3

Merged
hibryda merged 2 commits from docs/rmp-flags-config-reference into main 2026-06-02 22:49:59 +02:00
Owner

Adds the two RMP migration flags to docs/configuration.md (they were only in .env.example). New "Review module protocol" section: defaults, accepted values, module-load-read/restart-to-flip semantics, the startup review path: … log line, and the single-reviewer-only caveat for RMP_DRIVER. Doubles as the live-validation trigger for the RMP_DRIVER single-reviewer driver path.

Adds the two RMP migration flags to docs/configuration.md (they were only in .env.example). New "Review module protocol" section: defaults, accepted values, module-load-read/restart-to-flip semantics, the startup `review path: …` log line, and the single-reviewer-only caveat for RMP_DRIVER. Doubles as the live-validation trigger for the RMP_DRIVER single-reviewer driver path.
docs(config): document RMP_ENGINE + RMP_DRIVER in the configuration reference
All checks were successful
pr-reviewer CI / tsc --noEmit (pull_request) Successful in 28s
pr-reviewer CI / vitest (pull_request) Successful in 59s
pr-reviewer CI / tsc build (pull_request) Successful in 35s
c8fa845e51
The two RMP migration flags were documented only in .env.example. Add them to
docs/configuration.md (new "Review module protocol" section) with defaults,
accepted values, the module-load-read / restart-to-flip semantics, the
startup "review path: …" log line, and the single-reviewer-only caveat for
RMP_DRIVER (it takes effect only when REVIEWER_QUORUM_ENABLED is off).

Superseded by round 2.

Show previous round

hib-pr-reviewer review — PR #3 (hib/hib-pr-reviewer)

Round 1 — head c8fa845e51ef, base main, trigger opened

TL;DR: CONDITIONAL_APPROVE — 1 minor finding: the log-format hint covers only the RMP_DRIVER segment and omits the liveness-path segment that reflects RMP_ENGINE.

Summary

Documentation-only PR adding the RMP_ENGINE / RMP_DRIVER env-var table. The implementation claims were verified against src/worker.ts lines 1993–1994. One minor inaccuracy: the startup-log hint in the new prose (review path: …) covers only the RMP_DRIVER-controlled segment of the combined log line, leaving an operator with no clue how to verify RMP_ENGINE is active.

Blast Radius

Pure documentation change in docs/configuration.md. No code paths are modified; no runtime behaviour changes. Blast radius is limited to operator observability (knowing the right log field to look for).

BLAST_SCORE: 1/10

CI status (head c8fa845e51ef)

Overall: pending

3 checks: 3 pending

Check State Link
pr-reviewer CI / tsc --noEmit (pull_request) pending details
pr-reviewer CI / vitest (pull_request) pending details
pr-reviewer CI / tsc build (pull_request) pending details

Findings (1)

[MINOR] Log hint names only review path:, omitting liveness path: that reflects RMP_ENGINE

docs/configuration.md:50

Line 50 of the new section reads:

The engine startup log prints the active path (review path: …) so an operator can confirm which is live.

The actual log statement (src/worker.ts lines 1993–1994) emits a single combined line:

[hib-pr-reviewer] liveness path: … | review path: …

liveness path: reflects RMP_ENGINE; review path: reflects RMP_DRIVER. By surfacing only review path:, the doc gives an operator no hint for confirming RMP_ENGINE=1 is actually active.

Suggested fix:

-The engine startup log prints the active path (`review path: …`) so an operator can confirm which is live.
+The engine startup log prints both active paths (`liveness path: … | review path: …`) so an operator can confirm which is live for each flag.

Verdict

CONDITIONAL_APPROVE


hib-pr-reviewer • round 1 • 1 finding (1m) • 2026-06-02T20:42:56.286Z → 2026-06-02T20:44:27.707Z • posted-as: pr-reviewer-bot • model: auto

<!-- hib-pr-reviewer collapsed --> > _Superseded by round 2._ <details> <summary>Show previous round</summary> <!-- hib-pr-reviewer round:1 --> ## hib-pr-reviewer review — PR #3 (hib/hib-pr-reviewer) **Round 1** — head `c8fa845e51ef`, base `main`, trigger `opened` **TL;DR:** CONDITIONAL_APPROVE — 1 minor finding: the log-format hint covers only the RMP_DRIVER segment and omits the liveness-path segment that reflects RMP_ENGINE. ### Summary Documentation-only PR adding the `RMP_ENGINE` / `RMP_DRIVER` env-var table. The implementation claims were verified against `src/worker.ts` lines 1993–1994. One minor inaccuracy: the startup-log hint in the new prose (`review path: …`) covers only the `RMP_DRIVER`-controlled segment of the combined log line, leaving an operator with no clue how to verify `RMP_ENGINE` is active. ### Blast Radius Pure documentation change in docs/configuration.md. No code paths are modified; no runtime behaviour changes. Blast radius is limited to operator observability (knowing the right log field to look for). **BLAST_SCORE: 1/10** ### CI status (head `c8fa845e51ef`) **Overall: ⏳ pending** 3 checks: 3 pending | Check | State | Link | |---|---|---| | pr-reviewer CI / tsc --noEmit (pull_request) | ⏳ pending | [details](/hib/hib-pr-reviewer/actions/runs/56/jobs/0) | | pr-reviewer CI / vitest (pull_request) | ⏳ pending | [details](/hib/hib-pr-reviewer/actions/runs/56/jobs/1) | | pr-reviewer CI / tsc build (pull_request) | ⏳ pending | [details](/hib/hib-pr-reviewer/actions/runs/56/jobs/2) | ### Findings (1) #### **[MINOR]** Log hint names only `review path:`, omitting `liveness path:` that reflects RMP_ENGINE _docs/configuration.md:50_ Line 50 of the new section reads: > The engine startup log prints the active path (`review path: …`) so an operator can confirm which is live. The actual log statement (`src/worker.ts` lines 1993–1994) emits a single combined line: ``` [hib-pr-reviewer] liveness path: … | review path: … ``` `liveness path:` reflects `RMP_ENGINE`; `review path:` reflects `RMP_DRIVER`. By surfacing only `review path:`, the doc gives an operator no hint for confirming `RMP_ENGINE=1` is actually active. Suggested fix: ```diff -The engine startup log prints the active path (`review path: …`) so an operator can confirm which is live. +The engine startup log prints both active paths (`liveness path: … | review path: …`) so an operator can confirm which is live for each flag. ``` ### Verdict **CONDITIONAL_APPROVE** --- <sub>hib-pr-reviewer • round 1 • 1 finding (1m) • 2026-06-02T20:42:56.286Z → 2026-06-02T20:44:27.707Z • posted-as: pr-reviewer-bot • model: auto</sub> </details>
docs(config): name both startup log segments (review-finding fix)
All checks were successful
pr-reviewer CI / tsc --noEmit (pull_request) Successful in 21s
pr-reviewer CI / vitest (pull_request) Successful in 58s
pr-reviewer CI / tsc build (pull_request) Successful in 34s
bebba45bb4
Round-1 reviewer correctly flagged that the prose named only `review path:`
(RMP_DRIVER) while the actual startup line also prints `liveness path:`
(RMP_ENGINE). Name both segments so an operator can confirm each flag.

hib-pr-reviewer review — PR #3 (hib/hib-pr-reviewer)

Round 2 — head bebba45bb466, base main, trigger synchronize

TL;DR: NO_NEW_FINDINGS — prior round-1 condition (missing liveness path: attribution) is resolved in this synchronize push.

Summary

Round 1 finding is fixed. Line 50 now correctly attributes both log fields: `liveness path:` reflects `RMP_ENGINE` and `review path:` reflects `RMP_DRIVER`. No new actionable issues in the diff.

Blast Radius

Change is purely additive documentation in a single markdown file. No source code, config, or runtime behaviour is modified.

BLAST_SCORE: 1/10

CI status (head bebba45bb466)

Overall: pending

3 checks: 3 pending

Check State Link
pr-reviewer CI / vitest (pull_request) pending details
pr-reviewer CI / tsc build (pull_request) pending details
pr-reviewer CI / tsc --noEmit (pull_request) pending details

Findings

No new findings this round.

Claude reviewed and found nothing actionable.

Verdict

NO_NEW_FINDINGS


hib-pr-reviewer • round 2 • 0 findings • 2026-06-02T20:46:10.612Z → 2026-06-02T20:46:24.197Z • posted-as: pr-reviewer-bot • model: auto

<!-- hib-pr-reviewer round:2 --> ## hib-pr-reviewer review — PR #3 (hib/hib-pr-reviewer) **Round 2** — head `bebba45bb466`, base `main`, trigger `synchronize` **TL;DR:** NO_NEW_FINDINGS — prior round-1 condition (missing `liveness path:` attribution) is resolved in this synchronize push. ### Summary Round 1 finding is fixed. Line 50 now correctly attributes both log fields: `` `liveness path:` reflects `RMP_ENGINE` `` and `` `review path:` reflects `RMP_DRIVER` ``. No new actionable issues in the diff. ### Blast Radius Change is purely additive documentation in a single markdown file. No source code, config, or runtime behaviour is modified. **BLAST_SCORE: 1/10** ### CI status (head `bebba45bb466`) **Overall: ⏳ pending** 3 checks: 3 pending | Check | State | Link | |---|---|---| | pr-reviewer CI / vitest (pull_request) | ⏳ pending | [details](/hib/hib-pr-reviewer/actions/runs/57/jobs/1) | | pr-reviewer CI / tsc build (pull_request) | ⏳ pending | [details](/hib/hib-pr-reviewer/actions/runs/57/jobs/2) | | pr-reviewer CI / tsc --noEmit (pull_request) | ⏳ pending | [details](/hib/hib-pr-reviewer/actions/runs/57/jobs/0) | ### Findings **No new findings this round.** _Claude reviewed and found nothing actionable._ ### Verdict **NO_NEW_FINDINGS** --- <sub>hib-pr-reviewer • round 2 • 0 findings • 2026-06-02T20:46:10.612Z → 2026-06-02T20:46:24.197Z • posted-as: pr-reviewer-bot • model: auto</sub>
hibryda deleted branch docs/rmp-flags-config-reference 2026-06-02 22:49:59 +02:00
Sign in to join this conversation.
No reviewers
No labels
No milestone
No project
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
hib/hib-pr-reviewer!3
No description provided.