@1aboveio/engineering-skills 0.1.16 → 0.1.17

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.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@1aboveio/engineering-skills",
3
- "version": "0.1.16",
3
+ "version": "0.1.17",
4
4
  "description": "Issue-to-PR and release orchestrators, review, coverage, golden paths and test quality.",
5
5
  "license": "MIT",
6
6
  "repository": {
@@ -27,7 +27,7 @@
27
27
  "id": "engineering",
28
28
  "core": "@1aboveio/core-skills",
29
29
  "coreSkills": [],
30
- "digest": "2da4ff01960f2cd22b29fb4e0d7d3a3825c712b0b87f5b956f7d1bda09df32d1"
30
+ "digest": "eae585fe9a49dc577ca68ba48e060176008f7a2a6cd0e1f48e1b9f60fd12072d"
31
31
  },
32
32
  "publishConfig": {
33
33
  "registry": "https://registry.npmjs.org/",
@@ -4,7 +4,7 @@ description: "Slash/explicit-only (/better-review). Single-agent adversarial rev
4
4
  disable-model-invocation: true
5
5
  dependencies:
6
6
  - engineering-runtime
7
- release-version: 0.48.0
7
+ release-version: 0.48.1
8
8
  ---
9
9
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
10
10
  # Better Review
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  name: e2e-test
3
3
  description: "Playwright end-to-end guidance for affected browser journeys, render health and presentation checks, with route manifests and sweeps for broader audits. Use when browser routes or user flows ship or change, or a Playwright test flakes. NOT for non-browser tests, defining what to test (ensure-coverage), the curated ALIVE/DEAD golden-path gate (golden-path), or CI workflow YAML (github-actions)."
4
- release-version: 0.48.0
4
+ release-version: 0.48.1
5
5
  ---
6
6
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
7
7
  # E2E Testing Standard
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  name: engineering
3
3
  description: "Engineering conventions: feature-delivery priorities, focused verification and review, operational boundaries, and tracker routing. Standalone /engineering offers Global, Project, or Session setup and targeted resets for configured scopes; /engineering init keeps explicit project setup. Use when setting up AGENTS.md or CONTRIBUTING.md, saving or resetting engineering defaults, or before creating an engineering issue, spec, epic, ticket, or sub-issue on GitHub or Linear. Reads existing policy without writing during ordinary work. NOT the issue-to-PR workflow (resolve-issues), release execution (resolve-release), Git delivery (github-flow), or CI setup (github-actions)."
4
- release-version: 0.48.0
4
+ release-version: 0.48.1
5
5
  ---
6
6
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
7
7
  # Engineering
@@ -3,7 +3,7 @@ name: engineering-runtime
3
3
  description: Internal support package for shared Engineering Workflow executable behavior. Never select or invoke this package as a user capability.
4
4
  disable-model-invocation: true
5
5
  user-invocable: false
6
- release-version: 0.48.0
6
+ release-version: 0.48.1
7
7
  ---
8
8
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
9
9
  # Engineering Runtime
@@ -369,7 +369,7 @@
369
369
  "repository": "1aboveio/ai",
370
370
  "revision": "main"
371
371
  },
372
- "contentDigest": "3204ec23d0c79b0a5df8af2cf3a68a3bf839087b29969c74f5258ca8d156571b",
372
+ "contentDigest": "0ee641d7c1c90df5d1ca4d4f516f3aec0236ae224f2cfdf5e69a8d0023d0224f",
373
373
  "digestExcludes": []
374
374
  },
375
375
  {
@@ -382,7 +382,7 @@
382
382
  "repository": "1aboveio/ai",
383
383
  "revision": "main"
384
384
  },
385
- "contentDigest": "6e0d37af1ed2c9e38958d8ebb4f3c021491c2a8dd2fd12304d4b28159949b3c0",
385
+ "contentDigest": "b0c6b061b9d99a3fdf61ee41a3380986f8b0e6f687fc14158071ba5cfc4ad231",
386
386
  "digestExcludes": []
387
387
  },
388
388
  {
@@ -395,7 +395,7 @@
395
395
  "repository": "1aboveio/ai",
396
396
  "revision": "main"
397
397
  },
398
- "contentDigest": "a551b109b523dfd00d67402b38f130014207be98d2ccf524a768add85ab85911",
398
+ "contentDigest": "00afe10ecfc7fb382a3c9ddf7cfb1264eeeb003e935c3fd8eee09af7f0e6dc4c",
399
399
  "digestExcludes": []
400
400
  },
401
401
  {
@@ -408,7 +408,7 @@
408
408
  "repository": "1aboveio/ai",
409
409
  "revision": "main"
410
410
  },
411
- "contentDigest": "0d0a49fd2245eea13d6b994457e64450f7cec96b920c525138905fcfba5e2f44",
411
+ "contentDigest": "807efefc218519809541349ebc10358d37c5688d5d657fe2a58dc85e2a9aa055",
412
412
  "digestExcludes": []
413
413
  },
414
414
  {
@@ -421,7 +421,7 @@
421
421
  "repository": "1aboveio/ai",
422
422
  "revision": "main"
423
423
  },
424
- "contentDigest": "572b27835ce6883e3fbd8f7db6ba29a1c3a2e3a1217367bc08bf8cdce1f0ae9a",
424
+ "contentDigest": "efb97f0e66dec6c2d6894a1feb128892f6974d9aa71c83a29acabf22852f2228",
425
425
  "digestExcludes": []
426
426
  },
427
427
  {
@@ -434,7 +434,7 @@
434
434
  "repository": "1aboveio/ai",
435
435
  "revision": "main"
436
436
  },
437
- "contentDigest": "303f82fb4fd519446ecbb5b599ab179bec7ec74b2d160c0456ff841b97ff807d",
437
+ "contentDigest": "a54c0d8d50bd9de11b98118072a87f1fefb1930e4936eae4ad1c8d369125e80b",
438
438
  "digestExcludes": []
439
439
  },
440
440
  {
@@ -447,7 +447,7 @@
447
447
  "repository": "1aboveio/ai",
448
448
  "revision": "main"
449
449
  },
450
- "contentDigest": "b0823ca9b218d171039a5d86b3038afe6cbc9a0be96848e060ef6c198882ae5a",
450
+ "contentDigest": "2700f926acbfc82dec992a81f42373f5ac03d8e6cdfd9288e447e72e4215ca11",
451
451
  "digestExcludes": []
452
452
  },
453
453
  {
@@ -460,7 +460,7 @@
460
460
  "repository": "1aboveio/ai",
461
461
  "revision": "main"
462
462
  },
463
- "contentDigest": "90ae8a3116dff8510a5511ef58a15d6bd632eaa61ad192911399c8d2bd0c56a3",
463
+ "contentDigest": "a00c2848f686ee2d4775661f87c2569b4810480233f4a0649ca995f0bebdb2a8",
464
464
  "digestExcludes": []
465
465
  },
466
466
  {
@@ -473,7 +473,7 @@
473
473
  "repository": "1aboveio/ai",
474
474
  "revision": "main"
475
475
  },
476
- "contentDigest": "13df8c9fbfaeca374d1eccf278026515d3cefe05092693ad3789351982271c38",
476
+ "contentDigest": "27b87c7ac54bfc217ffe5c68fae0a9d9b1dd8619286c470ca721200bb6962bd8",
477
477
  "digestExcludes": []
478
478
  },
479
479
  {
@@ -486,7 +486,7 @@
486
486
  "repository": "1aboveio/ai",
487
487
  "revision": "main"
488
488
  },
489
- "contentDigest": "36fcee2a585b71c6b391a8d44a9378503f35715c83cd487c4b9e6fee35300ab5",
489
+ "contentDigest": "a67d55dda1c9278b16d192efe28d2827d2a1010b7c14bde8bf3ac10ce78adf13",
490
490
  "digestExcludes": []
491
491
  },
492
492
  {
@@ -499,7 +499,7 @@
499
499
  "repository": "1aboveio/ai",
500
500
  "revision": "main"
501
501
  },
502
- "contentDigest": "801d2c483d76803864eb4cb34e6e0981e1039eba4a9533e94ad6f9075416cb73",
502
+ "contentDigest": "15626fe123fbf537b89af3cb34349430fc1c76c4ad04e8853f6e761c7f45db30",
503
503
  "digestExcludes": []
504
504
  },
505
505
  {
@@ -512,13 +512,13 @@
512
512
  "repository": "1aboveio/ai",
513
513
  "revision": "main"
514
514
  },
515
- "contentDigest": "4cacea8ee67f05aabb5b16485e6093e6d2b5524ae25398fe28fefb8e6302bea7",
515
+ "contentDigest": "1bdafd5f8149b4020319869ae83c17734cec9c2c52f7dc28d5c672985ca73111",
516
516
  "digestExcludes": [
517
517
  "coherence/workflow.json"
518
518
  ]
519
519
  }
520
520
  ],
521
- "releaseIdentity": "9301ab0617c9f8d93c8e58cb6b37e3e3d487efdf24036c4c99a1defc08345e08",
521
+ "releaseIdentity": "446559a9feac423ab3c67e5b391ea300aeb1e5a80ec94a66a69062025e47a87a",
522
522
  "lifecycleAuthority": "native-skills-cli",
523
523
  "repairRecipe": {
524
524
  "id": "engineering-workflow-dependency-first",
@@ -2,7 +2,7 @@
2
2
  name: ensure-coverage
3
3
  description: "Define or audit behavior-first test criteria — ACs, required test levels, mock/integration/E2E policy — across three axes: breadth (Coverage Ledger), depth (over-mocking), presentation (perceivability of browser surfaces). NOT judging whether a single test is any good (test-quality), authoring browser tests (e2e-test), reviewing a PR (better-review), or delivering issues (resolve-issues)."
4
4
  disable-model-invocation: true
5
- release-version: 0.48.0
5
+ release-version: 0.48.1
6
6
  ---
7
7
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
8
8
  # Ensure Coverage
@@ -3,7 +3,7 @@ name: golden-path
3
3
  description: "Golden-path E2E: the short list of user journeys whose failure blocks a release, kept in a repo as one contract (golden-paths.json) with a generated GOLDEN_PATHS.md and a generated GitHub Actions workflow, and run against a local build or a deployed URL for an ALIVE/DEAD verdict. Use to set golden paths up in a repo ('init golden paths', 'add golden-path CI'), to migrate an older smoke.manifest.json or a hand-built smoke setup ('audit', 'migrate smoke'), to add, change, or remove a golden path, to ask 'is the build alive?' or run them before a PR or release, or when a change touches a flow listed in GOLDEN_PATHS.md. NOT for writing the Playwright specs themselves (e2e-test), deciding what a change must prove (ensure-coverage), or general workflow YAML (github-actions)."
4
4
  dependencies:
5
5
  - e2e-test
6
- release-version: 0.48.0
6
+ release-version: 0.48.1
7
7
  ---
8
8
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
9
9
  # Golden path
@@ -3,7 +3,7 @@ name: harness-runtime
3
3
  description: Internal import-only library shared by first-party engineering packages. Not a workflow skill; never select or invoke it as a user capability.
4
4
  disable-model-invocation: true
5
5
  user-invocable: false
6
- release-version: 0.48.0
6
+ release-version: 0.48.1
7
7
  ---
8
8
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
9
9
  # Harness Runtime
@@ -620,7 +620,8 @@ export function buildCatalogRequest({
620
620
  if (normalizedAvailable.available.length === 0) {
621
621
  return blocked(
622
622
  AGENT_CATALOG_BLOCK_CODES.CAPABILITIES_MISSING,
623
- 'no current-turn models were supplied; the harness must advertise its available models for this turn',
623
+ 'no current-turn models were supplied: the resolver input has no available[] entries; '
624
+ + 'collect verified host model and effort evidence before retrying; this does not establish that the harness lacks worker support',
624
625
  )
625
626
  }
626
627
  if (foregroundModel !== null && foregroundModel !== undefined && !nonEmptyString(foregroundModel)) {
@@ -1,6 +1,6 @@
1
1
  ---
2
2
  name: resolve-issues
3
- description: "Slash/explicit-only (/resolve-issues). Drive an issue, epic, or feature spec to reviewed, green PRs, one per issue, merged or handed back merge-ready: preflight, task map, exploration and a per-issue slice check, isolated implementers within live host capacity, then per issue its PR, independent better-review, CI, and queue/merge (or merge-ready hand-back) as soon as it is done. Asks once, at the end of preflight (target branch, merge mode, scheduler mode, plan) and resolves the worker roles' models as run metadata instead of asking, then never blocks on the human; small non-blocking problems collect in a dropbox, posted at the end as grouped ready-for-agent issues. /rush-issues is its no-questions shortcut. Do not auto-select. Use for whole-spec delivery, including high-risk work; risk changes model and validation depth, never the workflow. NOT for release work (resolve-release)."
3
+ description: "Slash/explicit-only (/resolve-issues). Drive an issue, epic, or feature spec to reviewed, green PRs, one per issue, merged or handed back merge-ready: preflight assessment with an explorer only for planning unknowns, a per-issue slice check, isolated implementers within live host capacity, then per issue its PR, independent better-review, CI, and queue/merge (or merge-ready hand-back) as soon as it is done. Asks once, at the end of preflight (target branch, merge mode, scheduler mode, plan) and resolves the worker roles' models as run metadata instead of asking, then never blocks on the human; small non-blocking problems collect in a dropbox, posted at the end as grouped ready-for-agent issues. /rush-issues is its no-questions shortcut. Do not auto-select. Use for whole-spec delivery, including high-risk work; risk changes model and validation depth, never the workflow. NOT for release work (resolve-release)."
4
4
  disable-model-invocation: true
5
5
  metadata:
6
6
  workflow-role: coordinator
@@ -8,7 +8,7 @@ dependencies:
8
8
  - better-review
9
9
  - golden-path
10
10
  - harness-runtime
11
- release-version: 0.48.0
11
+ release-version: 0.48.1
12
12
  ---
13
13
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
14
14
  # Resolve issues
@@ -54,7 +54,7 @@ after 2 fix rounds without a new low of open P0/P1 path rows; no fixed round cap
54
54
 
55
55
  ```text
56
56
  preflight: ready label -> model fills -> plan draft
57
- -> exploration -> size check (only scope or an open question holds a unit, with a split)
57
+ -> preflight assessment (optional explorer) -> size check (only scope or an open question holds a unit, with a split)
58
58
  -> one approval (target branch, merge mode, scheduler mode, plan)
59
59
  -> schedule runnable units within live host capacity:
60
60
  prepare a unit's workspace when its first worker is admitted (never all units up front)
@@ -77,9 +77,9 @@ Read each procedure right before its stage; never substitute your own.
77
77
 
78
78
  | Stage | Procedure |
79
79
  |---|---|
80
- | Preflight: ready label, models, plan draft, exploration, size check, approval; what any filed text may say | [references/preflight.md](references/preflight.md) ([Filed text](references/preflight.md#filed-text)) |
80
+ | Preflight: ready label, models, plan draft, assessment, size check, approval; what any filed text may say | [references/preflight.md](references/preflight.md) ([Filed text](references/preflight.md#filed-text)) |
81
81
  | Task plan shape and live board | [references/task-plan.md](references/task-plan.md) |
82
- | Exploration (preflight step 4) | [references/exploration.md](references/exploration.md) |
82
+ | Preflight assessment and optional exploration (step 4) | [references/exploration.md](references/exploration.md) |
83
83
  | Implementation loop and opening each PR | [references/implementation.md](references/implementation.md) |
84
84
  | Read-only warehouse canary | [references/canary.md](references/canary.md) |
85
85
  | Owned workspaces: plan, prepare, cleanup, recover, inventory | [references/workspaces.md](references/workspaces.md) |
@@ -105,7 +105,8 @@ Read each procedure right before its stage; never substitute your own.
105
105
  check. When a journey or case you run fails, copy its `NATIVE_FAILURE` line (or the failing step, error and
106
106
  evidence path) into the unit's plan row; an exit code alone is not a failure record.
107
107
  Leave code changes to workers so your attention stays on scheduling, and record each worker as the
108
- unit's `workerKind` (`implement`, `fix`, `review`). Exploration continues the run's one explorer. After a context compaction, read this file in
108
+ unit's `workerKind` (`implement`, `fix`, `review`). You own preflight findings;
109
+ delegate unresolved planning unknowns under [exploration.md](references/exploration.md#choose-the-owner). After a context compaction, read this file in
109
110
  full again before your next step: the summary keeps the run's state, not these rules. Fill
110
111
  free slots before starting your own work, stay within [expire.md](references/expire.md)'s limit, and send
111
112
  large output to files (read only the lines you need). Expect Pi's **Completions while busy: Steer active run**
@@ -149,12 +150,13 @@ Follow [preflight.md](references/preflight.md):
149
150
  efforts are run metadata, never a question.
150
151
  2. **Provisional target:** the default branch.
151
152
  3. **Plan draft:** one unit per issue, links, delivery mode; show the board.
152
- 4. **Exploration:** have one explorer record each unit's `slice` and,
153
- separately, its `proof`, in every run; reuse prior findings for units
153
+ 4. **Preflight assessment:** record each unit's `slice` and,
154
+ separately, its `proof`, yourself for clear work; use an explorer only
155
+ for concrete planning unknowns. Reuse prior findings for units
154
156
  whose evidence did not change
155
157
  ([exploration.md](references/exploration.md#reuse-first)).
156
158
  5. **Size check:** run `size-check.mjs` (exit 0; read `verdict`). On `hold` or `stop`, hold the listed
157
- units (their dependents wait, not started) and draft their splits (yours when none is drafted); on `exploration-unfinished`, continue the explorer.
159
+ units (their dependents wait, not started) and draft their splits (yours when none is drafted); on `exploration-unfinished`, complete the findings using step 4's owner routing.
158
160
  6. **Approval:** ask once for target, merge mode, scheduler mode, and plan;
159
161
  record the answers; plan each unit's branch, workspace, base, and lifecycle
160
162
  intent, and prepare a workspace only when its first worker is admitted
@@ -1,16 +1,44 @@
1
1
  # Exploration
2
2
 
3
- **One explorer per run, after the plan draft, records what sizes, orders, and
4
- blocks the units: behaviors, missing prerequisites, proof, line estimates,
5
- dependencies, contracts, and the questions that hold a unit.**
3
+ **Every run gets a coordinator-owned preflight assessment after the plan draft:**
4
+ behaviors, missing prerequisites, proof, line estimates, dependencies, contracts,
5
+ and the questions that hold a unit. A dedicated explorer is optional.
6
6
  In a one-unit run, leave `contracts`, `dependsOn`, and `stacksOn` empty.
7
7
 
8
- ## Spawn
8
+ ## Choose the owner
9
+
10
+ Read the issues, applicable repository guidance, and the few code or test paths
11
+ needed to check the plan. For clear, bounded work, write the findings yourself.
12
+ The findings are planning state, not product code; keep detailed code discovery
13
+ with implementers.
14
+
15
+ Delegate when a concrete unknown could change the plan: uncertain cross-unit
16
+ dependencies or contracts, unclear boundaries or likely splitting, or an unfamiliar
17
+ validation setup. Name the unknown and the affected units in the explorer's brief.
18
+ Multiple units or a high-risk label alone do not require an explorer; unresolved
19
+ planning uncertainty does. An unanswered product decision stays an open question,
20
+ not an invitation for the explorer to invent a requirement.
21
+
22
+ Use the same record and size check for both paths. Resolve each unit's planning
23
+ unknowns or record its open human decision before approval. On later scope or base
24
+ changes, apply this routing again only to affected units.
25
+
26
+ Resolve the assessment target to an exact commit before reading code: initially
27
+ the provisional target, then the approved target after a branch change. Record
28
+ that SHA as findings `base`; use it explicitly for reuse checks and any explorer
29
+ checkout. The coordinator's current branch is not evidence of the assessment target.
30
+
31
+ ## Optional explorer
32
+
33
+ Use at most one explorer at a time per run, on the fleet `explore` row, within
34
+ live host capacity. Reuse that explorer for further investigation when available.
35
+ Give it only the affected units and unknowns. While it writes findings, it owns
36
+ that file exclusively; collect its result before the coordinator updates it.
9
37
 
10
38
  | Field | Value |
11
39
  |---|---|
12
40
  | Model | The explorer fill ([preflight.md](preflight.md#1-model-slots)) |
13
- | Workspace | Read-only checkout of the provisional target; write only the findings directory |
41
+ | Workspace | Read-only checkout at the assessment target SHA; write only the findings directory |
14
42
  | Task plan | Absolute `task-plan.json` path; its unit ids key the findings |
15
43
  | Prior findings | Newest `.resolve-issues/rush/*/exploration/findings.json` for the same spec locator, if any |
16
44
  | Deadline | 15 minutes, plus 1 minute per unit beyond 4 |
@@ -18,15 +46,23 @@ In a one-unit run, leave `contracts`, `dependsOn`, and `stacksOn` empty.
18
46
 
19
47
  ## Reuse first
20
48
 
21
- **When prior findings exist, explore only the units whose evidence changed.**
22
- Run, before spawning, and give the explorer its output:
49
+ **Reuse requires unchanged requirements as well as unchanged code evidence.**
50
+ First compare each unit's current criteria, dependency contracts, relevant guidance,
51
+ and recorded decisions with the prior assessment. Reassess affected units even if
52
+ the helper below says `reuse`; it checks Git evidence paths, not those inputs.
53
+ When the prior inputs cannot be established, reassess that unit.
54
+
55
+ Run the code-evidence check against the exact assessment target, not the caller's
56
+ `HEAD`; include the output and any forced reassessments in an explorer brief:
23
57
 
24
58
  ```bash
25
59
  node <skill-base-dir>/scripts/findings-reuse.mjs \
26
- --prior <prior findings.json> --units <this run's unit ids, comma-separated>
60
+ --prior <prior findings.json> --units <this run's unit ids, comma-separated> \
61
+ --head <assessment-target-SHA>
27
62
  ```
28
63
 
29
- - `reuse`: copy the entry unchanged, with the new `base`.
64
+ - `reuse`: copy the entry unchanged only when the non-code inputs above also
65
+ match, with the new `base`.
30
66
  - `recheck`: re-explore the entry; its `changed` files are where to look.
31
67
  - `new`: explore from scratch.
32
68
 
@@ -34,7 +70,9 @@ Example: a resumed run starts after earlier units merged `src/charge/`; only
34
70
  the two units whose evidence names `src/charge/` are re-explored, the other
35
71
  four are copied.
36
72
 
37
- ## Brief
73
+ ## Assessment record
74
+
75
+ The coordinator and any delegated explorer use these rules:
38
76
 
39
77
  1. **Scaffold, then save one unit at a time.** Keep an existing
40
78
  `findings.json`; otherwise create it with `version: 2`, the `base` SHA,
@@ -106,7 +144,7 @@ four are copied.
106
144
  journey; `local` runs that Charge journey script on an owned emulator,
107
145
  not the repository's default phone journey.
108
146
  - Findings saved before `local` and `skill` existed are unfinished:
109
- continue the explorer to add them, even when reusing the rest.
147
+ add them before the size check, even when reusing the rest.
110
148
  - `physical`: evidence only a real device gives; else empty.
111
149
 
112
150
  - `groups` and `setups` (only when `slice.newProductBehaviorCount` is
@@ -193,10 +231,14 @@ it then.
193
231
 
194
232
  ## Completion
195
233
 
196
- Confirm the findings exist, the model read-back matches, and no product source
197
- changed. Never edit the findings yourself. Run the size check; on `exploration-unfinished` or
198
- tool failures, continue the explorer with the errors. At the deadline, have it
199
- save what it has and stop; continue it in the same directory for the units
200
- still empty while size-check problems plus empty units reach new lows, then follow
201
- [expire.md](expire.md). Never skip exploration;
202
- reuse is not skipping.
234
+ Confirm the findings exist and no product source changed. For delegated work,
235
+ also confirm the model read-back matches and inspect the result. Run the size
236
+ check; `exploration-unfinished` means complete or repair the findings, not spawn
237
+ a worker automatically. Correct a missing field yourself when the evidence is
238
+ clear; continue the explorer only for unresolved investigation.
239
+
240
+ At an explorer's deadline, have it save what it has and stop. Inspect its partial
241
+ findings, complete clear entries yourself, and continue investigation only for
242
+ remaining unknowns while size-check problems plus empty units reach new lows;
243
+ then follow [expire.md](expire.md). Finish the fleet `explore` row once the
244
+ assessment and size check are complete, even when no explorer was needed.
@@ -76,7 +76,7 @@ the same work. A passing setup check is not a substitute for the blocked behavio
76
76
  No additional report or result schema is required.
77
77
 
78
78
  Have the implementer treat findings as leads, find its own code, and report
79
- any `product`, `scope`, `design`, or `approval` question the explorer missed and any
79
+ any `product`, `scope`, `design`, or `approval` question the assessment missed and any
80
80
  human-only proof. Report each question in the result's `decisionNeeded`
81
81
  ([expire.md](expire.md#decide-before-you-defer)); add each proof as a
82
82
  handback step. Never edit `findings.json`.
@@ -11,10 +11,22 @@ until step 6.
11
11
  node <skill-base-dir>/scripts/plan-sync.mjs --find-run --runs .resolve-issues/rush --issue <owner/name#123>
12
12
  ```
13
13
 
14
- When it prints a directory, resume that run (`next-actions.mjs --run <dir> --resume` with
15
- [fresh capacity](implementation.md#live-capacity))
16
- instead of starting another; its plan picks up whatever the tree gained
17
- since. It exits 0 whether or not it found a run: no run prints nothing on stdout
14
+ When it prints a directory, reuse that run instead of starting another. Before
15
+ calling the scheduler, reload its plan, decisions and standing orders. Refresh the
16
+ source requirements as described under [issue-tree sync](task-plan.md#issue-tree-sync)
17
+ and resolve the recorded target to its current exact SHA. Apply
18
+ [assessment reuse](exploration.md#reuse-first) against those inputs.
19
+
20
+ If assessment was interrupted, findings are incomplete, or affected requirements or
21
+ target evidence changed, complete step 4 and rerun step 5 for those units before
22
+ scheduling them. Finish any outstanding step 6 approval once; preserve choices
23
+ already approved. A saved size stop remains held until reassessment clears it or
24
+ its `runWhole` recommendation is accepted. Do not treat a scheduler action as proof
25
+ that assessment or approval finished.
26
+
27
+ Then resume with `next-actions.mjs --run <dir> --resume` and
28
+ [fresh capacity](implementation.md#live-capacity); its plan picks up whatever the
29
+ tree gained since. The run finder exits 0 whether or not it found a run: no run prints nothing on stdout
18
30
  (`--json`: `{"found":false}`). Example: `/rush-issues #12` finds the run whose saved tree holds
19
31
  #12 and resumes it, never a second run beside it.
20
32
 
@@ -28,7 +40,7 @@ The script exits 0 whenever it ran (2 if it could not); read `outcome` in its JS
28
40
 
29
41
  - **`ready`:** continue. An open issue is work, whatever its sub-issues
30
42
  say: each `rollups` entry (open, every sub-issue closed) is a unit; the
31
- explorer records what is left, or that nothing is, and then the handback
43
+ assessment records what is left, or that nothing is, and then the handback
32
44
  says "close #N". Example: held #40's filed prerequisite #41 closed;
33
45
  #40 is a rollup and stays a unit.
34
46
  - **`unlabeled`:** continue as on `ready`, and record the `missing` issues for
@@ -78,9 +90,11 @@ posts them; they join this run's end-of-run review
78
90
 
79
91
  ## 1. Model slots
80
92
 
81
- Write this turn's host evidence to a temporary JSON file and run the fast path
82
- ([shared-modules.md](shared-modules.md#model-slots)); record its fills by exact id
83
- and effort. The skill names the roles — implementer, explorer, diagnoser,
93
+ Follow [model slots](shared-modules.md#model-slots): on Pi, inspect the live native
94
+ catalog first after checking task overrides and authoritative host preferences;
95
+ use the portable resolver when native role resolution is unavailable or cannot
96
+ validate either exact request. Record the actual fills by
97
+ exact id and effort. The skill names the roles — implementer, explorer, diagnoser,
84
98
  reviewer — and the harness agent catalog chooses the models, so rank nothing and
85
99
  copy nothing from a ladder. The fills are run metadata: show each role's model
86
100
  and effort on the board and never ask which model should implement. A model the
@@ -92,14 +106,23 @@ add the repository's deterministic domain checks and review depth, never change
92
106
  the workflow or re-choose the model.
93
107
 
94
108
  Pass a pinned implementer (`implementer=<id[@effort]>` in the invocation) as
95
- `--implementer`. Any `ok: false` result ends the run at preflight with its `code`
96
- and `reason`, starting nothing: `pinned-implementer-unavailable` means the pinned
97
- model is unavailable and no other implementer is substituted, and
98
- `catalog-unavailable` or `capabilities-missing` means this host cannot resolve
99
- roles yet. Never revive the old picker or reuse an earlier run's pick.
109
+ `--implementer` on the portable path. A declined result stops preflight with its
110
+ `code` and `reason`, starting nothing: `pinned-implementer-unavailable` means the
111
+ pinned model is unavailable and no other implementer is substituted;
112
+ `catalog-unavailable` means the portable resolver could not be invoked.
113
+ `capabilities-missing` means its input lacks model evidence: repair an omitted
114
+ field from current host evidence and retry, or report precisely what could not
115
+ be obtained. It does not establish a missing harness capability. Never revive
116
+ the old picker or reuse an earlier run's pick.
100
117
  Example: `/rush-issues 1aboveio/ai#1545 implementer=gpt-6-luna@high` on a host
101
118
  whose catalog does not offer `gpt-6-luna` stops at preflight and starts nothing.
102
119
 
120
+ **Catalog limitation:** optional exploration skips the worker, not catalog
121
+ resolution. Both paths still require all four fills, including
122
+ the explorer, even for coordinator-only assessment. Report a missing explorer as
123
+ a catalog block; do not extract other fills from a declined response or substitute
124
+ a model. Explorer-independent resolution needs a supported upstream protocol.
125
+
103
126
  ## 2. Provisional target
104
127
 
105
128
  Fetch remote heads; record the default branch as `targetBranch`.
@@ -118,11 +141,13 @@ Example: epic #12's #20 (signed release and go/no-go) and #21
118
141
  (physical acceptance) stay out of the plan; the run ends when its code units
119
142
  land instead of waiting on them.
120
143
 
121
- ## 4. Explore
144
+ ## 4. Assess the plan
122
145
 
123
- Run [exploration.md](exploration.md) on the provisional target, with any
124
- prior findings for the same spec. On tool failures or a size-check `exploration-unfinished` verdict,
125
- continue the explorer with the errors. Add the links the findings show.
146
+ Follow [exploration.md](exploration.md#choose-the-owner) on the provisional
147
+ target, reusing prior findings for the same spec. Write findings yourself for
148
+ clear work; delegate concrete planning unknowns to an optional explorer.
149
+ On tool failures or `exploration-unfinished`, complete the affected findings
150
+ using the same owner routing. Add the links the findings show.
126
151
 
127
152
  ## 5. Size check
128
153
 
@@ -154,7 +179,7 @@ by row 1 of [Never block on the human](#never-block-on-the-human).
154
179
  | `ok`, `warn` | Continue |
155
180
  | `hold` | Hold the `held` units (their dependents wait, not started); start the `runnable` units |
156
181
  | `stop` | Run-wide failure or every unit held; draft the held units' splits |
157
- | `exploration-unfinished` | Continue the explorer with the error |
182
+ | `exploration-unfinished` | Complete the findings using step 4's owner routing |
158
183
  | exit 2 | Fix the plan or command |
159
184
 
160
185
  Only scope and open questions hold a unit; proof only warns. Add each
@@ -285,7 +310,7 @@ Example: `/resolve-issues` answered "run the rest" files nothing;
285
310
  prerequisites keeps its unit, now depending on them.
286
311
  Example: #30 splits into #50 and #51, so the sync adds both, drops 30,
287
312
  and unit 34, `blocked_by` #30, now waits on 50 and 51.
288
- 4. Continue the explorer for the new units only; rerun the size check.
313
+ 4. Assess the new units only using step 4's owner routing; rerun the size check.
289
314
  5. Split a still-held slice again only while splitting makes progress: its
290
315
  largest slice is smaller than the one before.
291
316
  Example: a 7,287-line unit split into 3,455 and 2,150 lines; a held
@@ -348,16 +373,16 @@ in one interaction, with the native single-choice tool when there is one:
348
373
 
349
374
  Record each answer in `decisions`, then:
350
375
 
351
- - New target: fetch it, continue the explorer to recheck the findings, rerun
376
+ - New target: fetch it, reassess affected findings using step 4's owner routing, rerun
352
377
  the size check; ask again only if it worsens to `hold` or `stop`.
353
- - Trim: drop the scope, continue the explorer to remove its questions and
378
+ - Trim: drop the scope, remove its findings, questions and
354
379
  handback items, rerun the size check.
355
380
  - File the splits: follow [File a split](#file-a-split); ask again only if
356
381
  the size check returns `stop`.
357
382
  - Move to later: mark the units `blocked` (`blockedReason: human-decision`, `humanItem: 1`, the human's words as `humanNeed`); they stay in
358
383
  this run's plan and resume with it (one run per spec). Never start a
359
384
  separate follow-up run for the same spec.
360
- - Answer: record it, continue the explorer to set the question
385
+ - Answer: record it, set the question
361
386
  `resolved: true`, rerun the size check.
362
387
  - Run the rest: mark held units `blocked` (`blockedReason: size-hold`) with the split path;
363
388
  their dependents stay not started (`pending`).
@@ -11,7 +11,8 @@ restate its rules in the brief beyond the two below.
11
11
  - Reviewer model: a different id from the implementer's
12
12
  ([preflight.md](preflight.md#1-model-slots)). Launch the reviewer with the
13
13
  plan's recorded reviewer fill — its `launch`, `instructions`, `selection` and
14
- `catalogDigest` ([task-plan.md](task-plan.md#worker-selection-on-resume)) — as
14
+ native `catalogRevision` or portable `catalogDigest`
15
+ ([task-plan.md](task-plan.md#worker-selection-on-resume)) — as
15
16
  structured launch arguments, never as brief text and never a model id
16
17
  hardcoded here. On Pi that `launch` is what the live role resolver receives,
17
18
  `{ role }` unless an explicit task override or this host's own preference
@@ -37,9 +37,56 @@ Owner: the harness agent catalog. The workflow names four **roles** — implemen
37
37
  explorer, diagnoser (escalation), reviewer — and the harness decides which model
38
38
  fills each one and how that role is launched on this host. The shared bridge is
39
39
  [`harness-runtime/agent-catalog.mjs`](../../harness-runtime/agent-catalog.mjs);
40
- this skill owns no ladder, no tier ranking, and no model list. Write this turn's
41
- host evidence to a temporary JSON file first (any writable path; the run directory
42
- does not exist yet at step 1):
40
+ this skill owns no ladder, no tier ranking, and no model list.
41
+
42
+ ### Native Pi preflight
43
+
44
+ When Pi advertises `agents_catalog` and role-based spawning, use the live catalog
45
+ for a run with no task model/effort override or separately declared authoritative
46
+ host preference. Check those inputs before selecting this path: an unpinned task
47
+ can still have `hostPreferences`. The native catalog resolves roles against this
48
+ session's model registry and effort support; a second, caller-built inventory is
49
+ unnecessary.
50
+
51
+ 1. Call `agents_catalog` with `action: "list"`, then `action: "inspect"` for
52
+ `developer`, `explorer`, and `reviewer`. Inspect all three even when the explorer
53
+ will not run. The diagnoser uses the same developer role.
54
+ 2. Require each inspection's live `launchable: yes` and `availability: available`,
55
+ with its actual resolved model and effort. `definitionValid` and
56
+ `catalogLaunchable` alone establish only definition validity, not runtime
57
+ availability. A missing or blocked live decision stops preflight with that
58
+ diagnostic; saved defaults do not replace it.
59
+ 3. Record the catalog revision and each role's **actual** model and effort as run
60
+ metadata. A tier fallback can differ from the displayed default. Use the
61
+ developer result for implementer and diagnoser. Require the reviewer's actual
62
+ model id to differ from the implementer's, regardless of effort; otherwise
63
+ stop and report the conflict without choosing a replacement.
64
+ Save each fill in `selections` as `source: "native-catalog"`, `role`, `id`
65
+ (actual model), `key` (`id@effort` when effort is present), `effort` (actual
66
+ effort), `launch: { role }`, `instructions`
67
+ from the inspection, and the observed `catalogRevision`. Retain the native
68
+ inspection as `selection` evidence, including requested/actual values and
69
+ availability. This is native provenance, not a fabricated portable
70
+ `selection` object or `catalogDigest`.
71
+ 4. Record the later launch by role (`{ role: "developer" }`, `{ role: "explorer" }`, or
72
+ `{ role: "reviewer" }`), adding the task brief and authorized workspace/tool
73
+ controls. The native catalog applies its live instructions and defaults again
74
+ at admission. Record actual launch provenance if it differs from preflight.
75
+
76
+ This completes model preflight; continue to the provisional target without
77
+ running `preflight-models.mjs` or manufacturing `available[]`. Worker capacity is
78
+ a separate scheduling input, not model evidence. An explicit task model or effort
79
+ override or authoritative host model/effort preference still needs validation for
80
+ that exact request: use the portable path below when the native inspection cannot
81
+ validate it and preserve its structured launch fields. Keep the request; an
82
+ unresolved request does not authorize a default launch.
83
+
84
+ ### Portable preflight
85
+
86
+ For a host without native role resolution, or a task override or authoritative
87
+ host preference requiring portable validation, write observed host capabilities
88
+ to a temporary JSON file (any writable path; the run directory does not exist yet
89
+ at step 1):
43
90
 
44
91
  ```json
45
92
  { "available": [{ "model": "provider/model", "nativeModel": "host-id", "supportedEfforts": ["medium", "high"] }],
@@ -54,7 +101,7 @@ a field you cannot observe is left out, and the bridge never completes one.
54
101
 
55
102
  | Field | Meaning |
56
103
  |---|---|
57
- | `available[]` | This turn's models. `nativeModel` is required for a harness whose slots are launched by a bare host id (Codex); pass the ids the spawn primitive advertises, unchanged. `supportedEfforts` is the host's verified list per model. |
104
+ | `available[]` | This turn's models. Each entry needs verified `supportedEfforts`; IDs alone from configuration or `discover-models.mjs` do not supply effort support. `nativeModel` is required for a harness whose slots are launched by a bare host id (Codex); pass the advertised mapping unchanged. |
58
105
  | `launchCapabilities` | The launch controls this host exposes: `model`, `effort`, `instructions`, each a boolean. All three must be stated; omission is not "probably true". A native id and an effort say what the catalog chose, never that the adapter can set it. |
59
106
  | `hostPreferences` | This host's own role preferences, keyed by **role** (`developer`, `explorer`, `reviewer`), each `{ model?, effort? }`. Below a task override and above the role default: it is how a host says "this machine runs the developer role on this model". |
60
107
  | `fallbackToForeground` | Set `true` only when this host may use the foreground model as a fallback. Pi always may; another harness must opt in. Absent, no foreground fallback happens for it. |
@@ -115,6 +162,13 @@ pick, never fall back to the skill-owned catalog, and never let the reviewer sha
115
162
  the implementer's model id: the catalog enforces independence, and re-picking a
116
163
  replacement here would silently break it.
117
164
 
165
+ `capabilities-missing` describes incomplete resolver input, not proof that the
166
+ host cannot launch workers. Check the supplied file and this turn's native
167
+ capability source before declaring a harness limitation. Repair an omitted field
168
+ from verified evidence and retry preflight; if that evidence is unavailable,
169
+ report the exact missing input. Never derive effort ranges from a selected role
170
+ default or use free worker slots as a model inventory.
171
+
118
172
  **No fallback is invented.** The catalog substitutes only an available model from
119
173
  the same tier's configured `candidates`, and never for an explicit request: an
120
174
  unavailable pin or host preference blocks rather than sliding to a neighbour, and
@@ -1,6 +1,6 @@
1
1
  # Task plan
2
2
 
3
- Draft the plan in preflight, before exploration, and update it on every
3
+ Draft the plan in preflight, before the assessment, and update it on every
4
4
  worker and stage transition.
5
5
 
6
6
  ## Initialize the run
@@ -92,7 +92,7 @@ Fleet `status`: `pending`, `in-flight`, `succeeded`, `failed`,
92
92
  | `attempt`, `retries`, `escalations` | Attempt number; cycle counts; escalation-model attempts |
93
93
  | `worker`, `workerSince`, `clock` | Slot identifier of any worker running for the unit (implementer, reviewer, fix), `null` when none; when that worker started (ISO time, for [stuck-worker checks](expire.md#waits)); attempt and cumulative wall time |
94
94
  | `workerKind` | `implement`, `fix`, or `review`, set with `worker`; the [live-capacity scheduler](implementation.md#live-capacity) (`next-actions.mjs`) prints `worker-kind` for any other kind or none. An infrastructure diagnosis or a test-first worker is a `fix` worker. The explorer is the `explore` fleet row's worker, never a unit's; a merge-queue watcher is not a worker |
95
- | `estimatedLines` | Explorer estimate |
95
+ | `estimatedLines` | Preflight assessment estimate |
96
96
  | `sizeCheck` | Measures, `verdict` (the size check's per-unit verdict; `stop` means held), scope and proof verdicts, split or verification path; `runWhole: true` once a recommendation to run a held unit whole is accepted |
97
97
  | `note` | SHA, expiry reason code, failure and response, or blocker; for active debugging, the immediate acceptance checkpoint, baseline/result link and next distinguishing test ([delivery progress](../SKILL.md#delivery-progress)) |
98
98
  | `pr`, `headSha`, `reviewedHead` | PR URL, current HEAD (recorded from the implementer's first pushed commit on, so the draft PR opens early), last HEAD whose verdict was `pass` (reviewed or carried) |
@@ -123,10 +123,14 @@ rehydrate read it, and `next-actions.mjs` exits 2 without it), `planRevision`, `
123
123
  (`owner/name#N` or an issue URL; any other name turns off the issue-tree sync),
124
124
  exploration path, `decisions`, `standingOrders`, `schedulerMode` (`full`, `narrow`, or `off`), timestamps, and `excluded` (issues in the
125
125
  tree that are deliberately not units, each `{issue, reason}`: release or
126
- approval gates, and rollups the explorer found empty). A run resumed from an
126
+ approval gates, and rollups the assessment found empty). A run resumed from an
127
127
  older wave run keeps `merged` (unit ids merged by earlier runs).
128
128
 
129
- `selections` records each slot's fill as `preflight-models.mjs` returned it —
129
+ `selections` records each slot's fill from the selected
130
+ [model-preflight path](shared-modules.md#model-slots). A native Pi fill records
131
+ `source: "native-catalog"`, `role`, actual `id`, `key` and `effort`, `launch: { role }`,
132
+ inspected `instructions`, `catalogRevision` and the retained native inspection
133
+ as `selection`. A portable fill is what `preflight-models.mjs` returned —
130
134
  `id`, `key`, `effort`, `role`, `launch`, `instructions`, `selection`, and the
131
135
  `catalogDigest` it was resolved under — beside the explicit task or host override
132
136
  that produced it, when one did
@@ -162,10 +166,10 @@ to another launch or rewrite them.
162
166
  A plan saved before `selections` holds model ids and efforts and nothing saying
163
167
  who chose them. Read such a value as evidence of what ran, never as a pin: do
164
168
  not pass it as an explicit Pi model or effort. Before launching new workers,
165
- rerun `preflight-models.mjs` for this host ([preflight.md](preflight.md#1-model-slots))
169
+ repeat this host's model-preflight path ([preflight.md](preflight.md#1-model-slots))
166
170
  with only the overrides the run can evidence — a `standingOrders` line, a
167
171
  human-authored decision, or an explicit override in the invocation, never an
168
- agent recommendation or stored metadata alone — and record the fills it returns
172
+ agent recommendation or stored metadata alone — and record its resolved fills
169
173
  as `selections`. When an override's provenance, or a launch control the current
170
174
  host must advertise, cannot be established, block the new launch and name what
171
175
  is missing: report the code and reason and start nothing, rather than guessing a
@@ -173,6 +177,15 @@ pin. Workers already running or finished keep their configuration and history.
173
177
  A resume asks no model question; a human who changes one does so as a separate
174
178
  task input, recorded as a decision and resolved the same way.
175
179
 
180
+ For `source: "native-catalog"`, recheck current task overrides and authoritative
181
+ host preferences on resume. With neither, repeat live native inspection rather
182
+ than demanding a portable `catalogDigest` or running the portable script. A newly
183
+ declared preference or override returns through the exact-request validation path;
184
+ do not replay role-only defaults in its place. Keep
185
+ the old inspection as historical evidence; refresh preflight selection metadata
186
+ without rewriting any worker's actual launch history. A native inspection is not
187
+ an override, and its saved model and effort never become launch arguments.
188
+
176
189
  ### Standing orders
177
190
 
178
191
  **`standingOrders` are the run's only constraints.** A numbered list, one
@@ -212,14 +225,20 @@ Example: someone files #61 under epic #12 mid-run; the next resume prints
212
225
  - A failed fetch keeps the saved tree; a partial one never drops a unit or
213
226
  changes dependencies. A failed fetch before finishing never finishes:
214
227
  `next-actions.mjs` with [live capacity](implementation.md#live-capacity) prints `sync` until a fetch reads the tree.
215
- - `set-deps` only adds; a dependency the explorer recorded stays.
228
+ - At start and resume, fetch current spec and existing unit issue bodies from
229
+ their source and compare their criteria with the saved requirements, decisions,
230
+ contracts and relevant guidance. The tree fetch updates membership and links,
231
+ not issue bodies; unchanged sync output does not establish unchanged criteria.
232
+ Reassess affected units under [assessment reuse](exploration.md#reuse-first),
233
+ or reassess when the prior inputs cannot be established, before scheduling them.
234
+ - `set-deps` only adds; a dependency recorded in findings stays.
216
235
  - `cycle`: units depend on, or are chained on, each other, so none can move.
217
236
  Remove the link that is not a real dependency (on GitHub and in the plan);
218
237
  for a `stacksOn` loop, restack the unit it names onto the target.
219
238
  - `blocked-outside`: put the blocker into the tree (as a sibling of the
220
239
  waiting issue) or remove the link; never leave a unit waiting on an issue
221
240
  the run cannot see.
222
- - A rollup the explorer finds empty goes to `excluded` with "nothing left:
241
+ - A rollup the assessment finds empty goes to `excluded` with "nothing left:
223
242
  close it", and into the handback.
224
243
  - A unit that depends on an `excluded` issue waits for a human: it stays
225
244
  `pending` (not started), `next-actions.mjs` with [live capacity](implementation.md#live-capacity) gives it no action, and
@@ -230,7 +249,7 @@ Example: someone files #61 under epic #12 mid-run; the next resume prints
230
249
  ## Scope changes
231
250
 
232
251
  Record a new requirement as a decision, update the affected criteria,
233
- continue the explorer on affected and new units, rerun the size check, and
252
+ reassess affected and new units using [owner routing](exploration.md#choose-the-owner), rerun the size check, and
234
253
  show the board. Put the change in in-flight units' next brief; never copy
235
254
  files into a running workspace. Add new units as rows and schedule them
236
255
  normally.
@@ -313,7 +332,8 @@ text board. Keep the board and file in agreement. Never show an unsaved transiti
313
332
 
314
333
  ## Board
315
334
 
316
- Show the first board before the explorer spawn, as a record.
335
+ Show the first board before the preflight assessment, as a record. The fleet
336
+ `explore` row tracks that assessment, with a worker only when one is delegated.
317
337
 
318
338
  ```text
319
339
  resolve-issues rev 12 workers 2 elapsed 18m delivery stack
@@ -66,7 +66,7 @@ export function findingsReuse({ prior, units, head = 'HEAD' }) {
66
66
  let baseKnown = typeof prior?.base === 'string' && prior.base.trim() !== ''
67
67
  let changed = []
68
68
  if (baseKnown) {
69
- try { changed = git(['diff', '--name-only', prior.base, headSha]).split('\n').filter(Boolean) } catch { baseKnown = false }
69
+ try { changed = git(['diff', '--name-only', '--no-renames', prior.base, headSha]).split('\n').filter(Boolean) } catch { baseKnown = false }
70
70
  }
71
71
  const ids = units?.length ? units : Object.keys(prior?.units ?? {})
72
72
  return { base: prior?.base ?? null, head: headSha, ...decideReuse(prior, ids, changed, baseKnown) }
@@ -16,8 +16,8 @@
16
16
  // GitHub call. Sync after the run files an issue
17
17
  // unblock a unit marked blocked without a reason from BLOCK_REASONS
18
18
  // (`blockedReason`), or human-decision without `humanItem` (1-5) and
19
- // `humanNeed`: reopen it now; and on --resume, every blocked unit:
20
- // reopen it with fresh limits; its next step follows. A note never keeps a
19
+ // `humanNeed`: reopen it now; on --resume, reopen prior blocks with fresh
20
+ // limits except a current size stop without accepted runWhole. A note never keeps a
21
21
  // unit blocked. A complete review at the unit's HEAD that holds a P0/P1 but
22
22
  // wrote no fix brief and awaits no GitHub reviewer (a handback) gets a
23
23
  // fresh full round, so review-verdict hands it back again only if its
@@ -122,7 +122,7 @@ import { BLOCKING } from './review-verdict.mjs'
122
122
  export const WORKER_KINDS = Object.freeze(['implement', 'fix', 'review'])
123
123
  export const SKILL_PATH = join(dirname(dirname(fileURLToPath(import.meta.url))), 'SKILL.md')
124
124
  // Printed after every text call, and by rehydrate.mjs after a compaction.
125
- export const RULES = 'Per unit, run only implement, fix, and review workers, recorded as workerKind: no planning, architecture, product, or setup workers. Exploration continues the run\'s one explorer, on the fleet explore row. You coordinate: never edit product code. In full mode take every step from next-actions.mjs; in every mode the run ends only when it reports done. A HEAD after a passing review is carried; any other new HEAD takes the tier rereview-scope.mjs picks, never yours. Decide before you defer: a question a worker or review raises is answered and recorded, then the unit continues; retrying or deferring never answers it. Never block on the human: take the automatic answer and keep going. Fix the root cause, never only its symptom. Before each scheduling pass, refresh subagent_settings and pass its snapshot with --capacity-json; do not reuse a run-start budget or subtract workflow workers again.'
125
+ export const RULES = 'Per unit, run only implement, fix, and review workers, recorded as workerKind: no planning, architecture, product, or setup workers. You own preflight findings; use an optional explorer on the fleet explore row only for concrete planning unknowns (exploration.md). You coordinate: never edit product code. In full mode take every step from next-actions.mjs; in every mode the run ends only when it reports done. A HEAD after a passing review is carried; any other new HEAD takes the tier rereview-scope.mjs picks, never yours. Decide before you defer: a question a worker or review raises is answered and recorded, then the unit continues; retrying or deferring never answers it. Never block on the human: take the automatic answer and keep going. Fix the root cause, never only its symptom. Before each scheduling pass, refresh subagent_settings and pass its snapshot with --capacity-json; do not reuse a run-start budget or subtract workflow workers again.'
126
126
  export const DEFAULT_QUIET_MINUTES = 45
127
127
  // The reasons a unit may stay blocked (expire.md#reason-codes). Waiting on another unit, an
128
128
  // infrastructure or external blocker (a tree issue the unit waits on), and a size hold in a rush
@@ -205,15 +205,15 @@ export function nextActions(plan, reviews = {}, capacity, { resume = false, now,
205
205
  if (!Array.isArray(saved) || saved.length === 0) throw new NextActionsError('task plan has no units')
206
206
  const excluded = excludedIds(plan)
207
207
  const gates = (u) => (u.dependsOn ?? []).map(String).filter((d) => excluded.has(d))
208
- // In a rush run, a size-held unit is split, never deferred: it gets `split` until the sync
209
- // drops it (split into sub-issues) or a rerun size check clears it.
208
+ // A current size stop survives resume. Rush splits it; a normal run may hand it back.
210
209
  const shortcut = plan.skill === 'rush-issues'
211
- const sizeHeld = (u) => shortcut && (u.sizeCheck?.verdict ?? u.sizeCheck?.scope) === 'stop' && u.sizeCheck?.runWhole !== true && !codeDone(u)
210
+ const heldForSize = (u) => (u.sizeCheck?.verdict ?? u.sizeCheck?.scope) === 'stop' && u.sizeCheck?.runWhole !== true && !codeDone(u)
212
211
  && !hasWorker(u) && u.status !== 'succeeded' && u.status !== 'cancelled' && gates(u).length === 0
213
212
  && !excluded.has(String(u.id))
213
+ const sizeHeld = (u) => shortcut && heldForSize(u)
214
214
  const reasonOk = (u) => BLOCK_REASONS.includes(u.blockedReason) && !(shortcut && u.blockedReason === 'size-hold')
215
215
  && (u.blockedReason !== 'human-decision' || namesHumanItem(u))
216
- const reopen = (u) => (resume || !reasonOk(u)) && u.status === 'blocked' && !hasWorker(u) && gates(u).length === 0 && !sizeHeld(u)
216
+ const reopen = (u) => (resume || !reasonOk(u)) && u.status === 'blocked' && !hasWorker(u) && gates(u).length === 0 && !heldForSize(u)
217
217
  // `stack` and `chain` both open a child's PR once its parents' PRs are open; only the PR base differs.
218
218
  if (plan.deliveryMode !== undefined && !['stack', 'chain', 'wait'].includes(plan.deliveryMode)) throw new NextActionsError(`deliveryMode must be stack, chain, or wait: ${JSON.stringify(plan.deliveryMode)}`)
219
219
  const chain = plan.deliveryMode === 'chain'
@@ -139,7 +139,7 @@ export function planSync(plan, tree) {
139
139
  const { deps } = wants(key)
140
140
  const id = idFor(key, repo)
141
141
  const rollup = (node.children ?? []).length > 0
142
- add('add-unit', id, `add unit ${id} for ${key} "${node.title}"${deps.length ? ` with dependsOn [${deps.join(', ')}]` : ''}: fetch its issue into the run, continue the explorer for it, rerun the size check${rollup ? '; every sub-issue is closed, so the explorer records what is left or that nothing is (then hand back "close it")' : ''}`, { issue: key, dependsOn: deps })
142
+ add('add-unit', id, `add unit ${id} for ${key} "${node.title}"${deps.length ? ` with dependsOn [${deps.join(', ')}]` : ''}: fetch its issue into the run, assess it using exploration.md's owner routing, rerun the size check${rollup ? '; every sub-issue is closed, so record what is left or that nothing is (then hand back "close it")' : ''}`, { issue: key, dependsOn: deps })
143
143
  }
144
144
  for (const [key, u] of byKey) {
145
145
  if (DONE.has(u.status)) continue
@@ -30,7 +30,8 @@
30
30
  // (ok, warn, hold, stop, or exploration-unfinished); 2 for a bad plan or command.
31
31
  // `exploration-unfinished` (printed as `{ verdict, reason }`) means findings that are unreadable,
32
32
  // not version 2, or missing a planned unit, its `slice`, its `proof`, or its `estimatedLines`;
33
- // only the explorer fixes it. --exit-verdict restores the old exit codes for one release
33
+ // complete the preflight findings using exploration.md's owner routing.
34
+ // --exit-verdict restores the old exit codes for one release
34
35
  // (deprecated): 0 for ok/warn, 4 for hold, 1 for stop, 3 for exploration-unfinished.
35
36
 
36
37
  import { readFileSync } from 'node:fs'
@@ -78,7 +79,7 @@ export const FINDINGS_VERSION = 2
78
79
 
79
80
  export class SizeCheckError extends Error {}
80
81
 
81
- // Unfinished exploration: the explorer, not the orchestrator, fixes it.
82
+ // Unfinished assessment; keep the exported error and verdict compatible with saved runs.
82
83
  export class ExplorationUnfinished extends SizeCheckError {}
83
84
 
84
85
  function unitsOf(plan) {
@@ -262,9 +263,9 @@ function environmentChecks(units, byUnit) {
262
263
  // ---- The whole check ----
263
264
 
264
265
  // A unit counts as explored once its entry carries a valid `slice`, `proof`,
265
- // and `estimatedLines`; the explorer's scaffold (an empty entry per unit) does not pass.
266
+ // and `estimatedLines`; a scaffold (an empty entry per unit) does not pass.
266
267
  function findingsUnitsOf(findings, units) {
267
- const unfinished = (why) => new ExplorationUnfinished(`exploration unfinished: ${why}; continue the explorer`)
268
+ const unfinished = (why) => new ExplorationUnfinished(`exploration unfinished: ${why}; complete the preflight findings`)
268
269
  if (findings?.version !== FINDINGS_VERSION) throw unfinished(`findings version is ${JSON.stringify(findings?.version)}, expected ${FINDINGS_VERSION}`)
269
270
  const byUnit = findings.units
270
271
  if (!byUnit || typeof byUnit !== 'object' || Array.isArray(byUnit)) throw unfinished('findings has no units map')
@@ -481,7 +482,7 @@ if (isMainModule(import.meta.url)) {
481
482
  process.exitCode = opts.exitVerdict ? EXIT[result.verdict] : 0
482
483
  } catch (error) {
483
484
  if (error instanceof ExplorationUnfinished) {
484
- // Unfinished exploration is an answer (the explorer continues), not a failure of the script.
485
+ // Incomplete findings are an answer, not a failure of the script.
485
486
  process.stdout.write(`${JSON.stringify({ verdict: 'exploration-unfinished', reason: error.message }, null, 2)}\n`)
486
487
  process.exitCode = process.argv.includes('--exit-verdict') ? 3 : 0
487
488
  } else {
@@ -4,7 +4,7 @@ description: "Slash/explicit-only (/resolve-release). Full GitHub Flow release f
4
4
  disable-model-invocation: true
5
5
  metadata:
6
6
  workflow-role: coordinator
7
- release-version: 0.48.0
7
+ release-version: 0.48.1
8
8
  ---
9
9
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
10
10
  # Resolve release
@@ -6,7 +6,7 @@ metadata:
6
6
  workflow-role: coordinator
7
7
  dependencies:
8
8
  - resolve-issues
9
- release-version: 0.48.0
9
+ release-version: 0.48.1
10
10
  ---
11
11
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
12
12
  # Rush issues (shortcut of resolve-issues)
@@ -6,7 +6,7 @@ metadata:
6
6
  workflow-role: coordinator
7
7
  dependencies:
8
8
  - resolve-release
9
- release-version: 0.48.0
9
+ release-version: 0.48.1
10
10
  ---
11
11
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
12
12
  # Rush release (shortcut of resolve-release)
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  name: test-quality
3
3
  description: "Seven principles for tests that catch real bugs, and an audit that reports weak, redundant, and not-coverage tests without fixing them. Use when writing, changing, or reviewing tests, or when asked whether tests are useful, redundant, over-mocked, flaky, only assert static text, or test config files, or asked to make a slow test suite faster without losing coverage, or when the user asks to write the testing conventions into a repo's CONTRIBUTING.md. NOT defining what a spec must prove (ensure-coverage) or authoring browser tests (e2e-test)."
4
- release-version: 0.48.0
4
+ release-version: 0.48.1
5
5
  ---
6
6
  **Updates:** Once/session run `node <skill-base-dir>/release-check.mjs`.
7
7
  # Test Quality