@brainervirus/workit-opencode 8.23.0 → 8.24.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -40,7 +40,7 @@ units of ~100+ lines each. Coupled work stays with one owner. Workit never spawn
40
40
  chain resumes. A fix round and recheck that both fail: a decider, not a
41
41
  third round; product choices go to the user (`references/orchestration.md`).
42
42
  7. **Verify by the workspace `verification` setting** (`workit grant show`).
43
- A report is a pointer, never evidence. `self`: a verifier that wrote none
43
+ A report is a pointer, never evidence. `proportional` (default): each slice's lane (`workit pr status`) decides; none needs no verdict, self the author's `--self`, focused (cheap model: Claude Code `model: sonnet`) and full a verifier after the last push. `self`: a verifier that wrote none
44
44
  of the slices, or you if you wrote none. `independent` or high risk: a
45
45
  separate verifier session, never yours (doctrine: the ledger only refuses
46
46
  authors). One verifier may take a batch of slices, one verdict per branch
@@ -36,11 +36,11 @@ description: Build a requested change in small verified steps - follow local pat
36
36
  when that was requested, or the request implies delivery and the effective
37
37
  endpoint in `workit grant show` is `pr`, `green` or `merged`; `green` and
38
38
  `merged` keep babysitting. Otherwise it is `commit`.
39
- 7. Verify, as far as the policy asks. Normal risk: after `workit check test`
40
- passes, record your own `workit ledger verdict tests-verified --self --how
41
- "<what you ran>"`; it reads self-reviewed, never verified, and is enough.
42
- Ask a non-author only for high risk, a workspace with `verification:
43
- independent` (`workit grant show`) or a verified merge: a fresh verifier with its own host-derived session (Claude Code and Codex:
39
+ 7. Verify as far as the review lane asks (`workit policy assess`, `workit pr
40
+ status`). Lane none: no verdict. Lane self: after `workit check test` passes,
41
+ record `workit ledger verdict tests-verified --self --how "<what you ran>"`;
42
+ it reads self-reviewed, never verified, and is enough. Focused, full, high
43
+ risk or `verification: self|independent` merges: one fresh verifier per PR after the last push, its own host-derived session (Claude Code: the `verifier` agent, `model: sonnet` when focused; Codex:
44
44
  the `verifier` agent; OpenCode: a child session; Pi: a supervised worker),
45
45
  never a hand-minted id. Before saying done, reconcile
46
46
  every named deliverable against the target checkout and observe it (for a push:
@@ -13,6 +13,9 @@ OpenCode, a child session; Pi, a supervised worker. Never hand-mint one such
13
13
  as `<lead>-v1`: it is refused as `forged_session`.
14
14
  An author's `--self` verdict reads self-reviewed and never counts as
15
15
  independent; `type-check-only` never proves a behavior change.
16
+ Depth follows the review lane in `workit pr status`: focused reads the diff and
17
+ runs its tests; full also drives the real surface (verify-<app>). Review once
18
+ per PR after its last push; a second pass on an unchanged head is waste.
16
19
 
17
20
  1. Pin the candidate: `git rev-parse HEAD`, the base, `git diff <base>...HEAD`;
18
21
  for a PR, `workit pr status --json` (checks, unresolved threads).
@@ -51,8 +51,7 @@ the same check. Fix round and recheck both failed: a decider (call the skill too
51
51
  allows; no sleep loop. Red: read `logTail`, classify. Flake or infra: one
52
52
  `workit ci rerun --failed --reason flake|infra` per head. Real: reproduce
53
53
  with `workit check`, fix the root cause, batch every fix in one push.
54
- 6. **Verified.** High risk, `verification: independent` or a verified merge: a
55
- non-author verdict (call the skill tool with `workit-review`); else a self verdict stands. Land only when
54
+ 6. **Verified.** The lane in `workit pr status` decides: none, no verdict; self, your `--self` verdict; focused or full (or `verification: self|independent`), one non-author verifier per PR after the last push, never per push (call the skill tool with `workit-review`), on Claude Code `model: sonnet` when focused. Land only when
56
55
  granted: `workit stack land` or `workit pr merge`; `--unverified --reason` if user-asked.
57
56
  7. **Observe it landed:** `workit verify-delivery pr` or `merge`. After a failed
58
57
  verdict or a check red 3+ times, at any endpoint: offer `/wk-retro` in one