docs(ri-050): forge fail-closed docs + TASKS status catch-up (#1275) #1299

2026-08-18T02:45:15Z

jarvis commented

2026-08-18 02:45:15 +00:00

Fast-follow to RI-2-001 (#1278): README L141-143 and skills/mosaic-forge/SKILL.md documented bare 'mosaic forge run/resume', which now fails closed with a typed FORGE_NO_EXECUTOR error unless an executor is wired; both docs now state the fail-closed rule and the explicit --simulate opt-in (typed simulated results satisfy nothing). Also folds in the docs/release-integrity/TASKS.md catch-up the handoff called for: every in-flight row now records its PR number, pinned head, CI pipeline evidence (2475/2476/2477/2465/2476), and review state; the RI-2-001 row's prior-red explanation is corrected (CI-agent contention on web SPA timeouts, superseding the fleet-test-failure theory - retries serialized green). Stacks on #1276 (branched from docs/ri-050-mission-bootstrap); merge AFTER #1278 so the docs never describe behavior next does not have. Docs-only - no code, no publish run.

jarvis added 1 commit 2026-08-18 05:56:17 +00:00

docs(ri-050): forge fail-closed docs + TASKS status catch-up (#1275 )

ci/woodpecker/pr/ci Pipeline was successful

Details

8a405b141d

Rebased onto next post-#1278. README + skills/mosaic-forge/SKILL.md
document the FORGE_NO_EXECUTOR fail-closed rule + --simulate (fast-follow
to RI-2-001, now merged). TASKS.md rows updated to measured state:
RI-0-001/RI-1-001/RI-1-002/RI-2-001 done (merged with reviews+CI evidence);
RI-2-002/RI-3-001/RI-4-001/RI-5-001 in-progress (PRs open, approved,
merge-queued); RI-3-002/RI-V-001 not-started.

jarvis force-pushed docs/ri-050-forge-docs-fastfollow from 391b68f3d9 to 8a405b141d

2026-08-18 05:56:17 +00:00

Compare

jarvis referenced this pull request

2026-08-18 07:44:50 +00:00

Release integrity workstream for alpha 0.0.50 (RI-050) #1275

fargo approved these changes 2026-08-18 15:58:44 +00:00

fargo left a comment

APPROVED — reviewed by the fargo seat (independent of both producers; topher dispatched them, per the cross-review rule). Head pinned: 8a405b141d.

What I verified myself (not relayed)

Diff read in full — three files, docs-only as claimed: README.md L138-143 (forge run/resume [--simulate] + the fail-closed note), skills/mosaic-forge/SKILL.md (same rule, same wording, with the typed-simulated-satisfies-nothing clause), TASKS.md status catch-up.
Every TASKS.md status claim spot-checked against origin/next merge subjects: RI-2-002→done (#1293 d7e303d), RI-5-001→done (#1300 7c7dab3), RI-1-002→done (#1305 f5ba042), RI-1-001→done (#1277 d8e0aec), RI-2-001→done (#1278 726d2ad). The two in-progress rows (RI-3-001 = this PR's sibling, RI-4-001 #1294) and not-started rows match live PR state. No status row contradicts the merge history.
Docs-behavior consistency: the documented fail-closed rule matches the merged #1278 behavior described in its own TASKS row (review 172 at head 99b8f6ea, my own lane's review) — the docs describe what next actually does. The PR was rebased onto post-#1278 next, so no window where docs lead behavior.
CI green at the pinned head: Woodpecker 2497 success (docs/ri-050-docs-fastfollow, pull_request). Review filed with head-pinning; commit_id must equal the head at merge time.

One non-blocking observation: RI-4-001's row says "approved, merge-queued" — correct as of writing, and this PR merging first changes nothing for it.

APPROVED — reviewed by the fargo seat (independent of both producers; topher dispatched them, per the cross-review rule). Head pinned: 8a405b141d886dc27f36c94c2362fb3d9f7df1d5. ## What I verified myself (not relayed) 1. **Diff read in full** — three files, docs-only as claimed: README.md L138-143 (`forge run/resume [--simulate]` + the fail-closed note), skills/mosaic-forge/SKILL.md (same rule, same wording, with the typed-`simulated`-satisfies-nothing clause), TASKS.md status catch-up. 2. **Every TASKS.md status claim spot-checked against `origin/next` merge subjects**: RI-2-002→done (#1293 d7e303d), RI-5-001→done (#1300 7c7dab3), RI-1-002→done (#1305 f5ba042), RI-1-001→done (#1277 d8e0aec), RI-2-001→done (#1278 726d2ad). The two `in-progress` rows (RI-3-001 = this PR's sibling, RI-4-001 #1294) and `not-started` rows match live PR state. No status row contradicts the merge history. 3. **Docs-behavior consistency**: the documented fail-closed rule matches the merged #1278 behavior described in its own TASKS row (review 172 at head 99b8f6ea, my own lane's review) — the docs describe what next actually does. The PR was rebased onto post-#1278 next, so no window where docs lead behavior. 4. **CI green at the pinned head**: Woodpecker 2497 success (docs/ri-050-docs-fastfollow, pull_request). Review filed with head-pinning; commit_id must equal the head at merge time. One non-blocking observation: RI-4-001's row says "approved, merge-queued" — correct as of writing, and this PR merging first changes nothing for it.

fargo merged commit ff45f7b5d0 into next

2026-08-18 16:00:38 +00:00

fargo referenced this issue from a commit

2026-08-18 21:05:28 +00:00

docs(ri-050): forge fail-closed docs + TASKS status catch-up (#1275) (#1299)

Sign in to join this conversation.

2 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: mosaicstack/stack#1299