1 Commits
8 changed files with 58 additions and 38 deletions
+1
View File
@@ -179,6 +179,7 @@ creating thin wrapper checklist files would add no value.
| 8.5 | 防范范围收窄:明令禁止「只看重点/从简」类弱化措辞(IEEE 1028-2008 完整性) | ☐ | ☐ | |
| 8.6 | 留痕:审计产物落盘到结构化文件,而非仅存于对话 | ☐ | ☐ | |
| 8.7 | 闭环:存在重审循环与收敛/审批终止条件(ISO 19011:2018 跟踪与关闭) | ☐ | ☐ | |
| 8.8 | 流程自身测试为绿:Orchestrator(bash 可用)在派发审计员**之前**机械运行流程自身的测试套件与守卫门(`bun run test:parallel` 全量分片 + `check:*` 守卫族,如 core-parity / encoding-health / derived-artifacts),并把命令、基线 SHA 与结果摘要落盘到 `<runs-root>/audit/reviews/audit-process/process-tests-green.md`(Tier 1);审计员(只读,无 bash)核对该记录存在于当前基线且为绿——任何红测试/红守卫本身即流程缺陷,按 findings 登记(缺陷逃逸案例:红测试存活 3 轮审计、无任何维度运行过测试套件,[org-internal #4787]) | ☐ | ☐ | 运行归 Orchestrator、核对归审计员(角色边界:`core/skills/_shared/review-orchestrator-rules.md` §No bash for Explorer sub-agents) |
---
+1 -1
View File
@@ -1,6 +1,6 @@
# Release 检查清单
> 发版前自检。Release Manager 在 Phase 1 和 Phase 6 完整过一遍。
> 发版前自检。Producer (Worker) 在 Phase 1 和 Phase 6 完整过一遍。
> **维度命名空间(dimension namespace)**: 本清单中的维度代码(如 TST/BEH/GATE/REG)是**清单局部(checklist-local)**标识符,仅在本文档内唯一;其他清单可能对相同字母串绑定不同概念(如 code-review.md 的 TST=Test Quality、verification.md 的 TST=Test Suite Completeness、port.md 的 TST=Test Porting)。跨清单引用时必须带清单限定(如 code-review.md TST 5.8),不得使用裸代码。
@@ -324,8 +324,15 @@ After the {REVISION_ROLE} completes revision:
round, or the diff touches a frozen cross-session contract) —
`<instance-root>/workflow-routing.yaml` `review.max_rounds` ([org-internal #3567], process-diet
P0b). At budget without convergence: STOP, register the remaining
BLOCKER/MAJOR findings as TD registry rows, verdict FAIL if any BLOCKER
remains else PASS — the debt is carried by those TD registry rows
BLOCKER/MAJOR findings as TD registry rows, then compute the verdict
with the shared three-tier ladder (`core/schemas/synthesis.schema.json`
`overall_verdict` + `core/skills/_shared/review-orchestrator-rules.md`
rule 3): FAIL if any BLOCKER or FAIL dimension remains, else WARN,
PASS only if every remaining dimension is PASS ([org-internal #4786]:
the prior binary "else PASS" wording silently upgraded WARN-eligible
outcomes to PASS; a PASS round has already stopped at the convergence
table above, so the residual tail lands WARN or FAIL). The debt is
carried by those TD registry rows
(see `review-code/SKILL.md` Phase D). This
supersedes the 2026-08-25 [org-internal #3380] calibration (p50=2, p95=5, n=66): the runs
past round 3 are the marathon tail this budget exists to cut; their residue
+4 -4
View File
@@ -3,7 +3,7 @@ name: image-interpret
description: >
Use ONLY when the agent needs to read or analyze an image (screenshot,
design mockup, UI capture, .png/.jpg/.webp file) but the active model lacks
vision capability. Delegates to the Image Interpreter role (vision-capable
vision capability. Delegates to the Tool role (image interpretation) (vision-capable
model) which returns a structured text spec — layout regions, colors (hex),
spacing (px), typography, component hierarchy, and responsive breakpoints.
Do not use for non-image tasks or when the active model already supports
@@ -31,7 +31,7 @@ role: Tool
You (the calling agent) cannot read images — your model has no vision. Any
task that requires *seeing* a visual artifact must be delegated to the
**Image Interpreter** role, which runs on a vision-capable model and returns a
**Tool** role (image interpretation), which runs on a vision-capable model and returns a
structured text description you can act on.
Trigger situations:
@@ -64,7 +64,7 @@ Trigger situations:
)
```
4. **Consume the returned text spec.** The Image Interpreter returns a
4. **Consume the returned text spec.** The Tool sub-agent (image interpretation) returns a
structured description only — it cannot write code or artifacts. Use the
returned spec to drive your own work (code generation, design review,
planning).
@@ -74,7 +74,7 @@ Trigger situations:
## Notes
- The Image Interpreter is **read-only** — it describes images, it does not edit
- The Tool sub-agent (image interpretation) is **read-only** — it describes images, it does not edit
files or write code.
- One delegation per distinct visual question. Batch multiple images into a
single delegation only when they must be compared.
+31 -28
View File
@@ -1,12 +1,11 @@
---
name: release
description: >
Use ONLY when ready to cut a release. The Release Manager (Worker) inspects
Use ONLY when ready to cut a release. The Producer (Worker) inspects
project state (git log, version files, build) to determine the semver bump,
generate the changelog, tag, and produce a release artifact. Works on any
project state — no pipeline artifacts required.
triggers:
- release
- cut release
- 发版
- publish release
@@ -14,7 +13,6 @@ triggers:
role: Producer
---
> Core 中立版(Increment 6a 改写,原 deferHard verbatimDir)。机制、结构与 frontmatter 保持;实例术语(工具名、路径、工单号)按 `core/adapters/TERMINOLOGY.md` 绑定到具体实例。
# Release
Prepare and tag a release based on project facts — git history since the last
@@ -28,14 +26,14 @@ produced here is the input to project-specific deployment workflows.
## Agent Role
The release is owned and executed by the **Release Manager** (Worker).
The release is owned and executed by the **Producer** (Worker).
**Context compaction**: release is a pipeline stage boundary. The main session
compacts at this clean boundary ONLY when a capacity/projection trigger holds,
per `core/rules/compact.md` §"Stage-boundary compaction" (long multi-stage
per `.octopus/rules/compact.md` §"Stage-boundary compaction" (long multi-stage
runs — DAG Epic orchestration — keep the legacy every-boundary compaction;
short runs — bugfix / DAG task — and standalone runs default to NOT
compacting). The Release Manager itself is single-phase and
compacting). The Producer itself is single-phase and
artifact-driven (version files, changelog, tag); a mid-run compaction loses
nothing — re-read the release checklist state and git log to resume.
@@ -46,7 +44,7 @@ nothing — re-read the release checklist state and git log to resume.
Before starting the release:
- [ ] Working tree is clean (`git status` has no uncommitted changes).
- [ ] `core/checklists/release.md` is accessible.
- [ ] `.octopus/checklists/release.md` is accessible.
---
@@ -65,7 +63,7 @@ Run each check against the live project. Stop and report failures.
- No new HIGH or CRITICAL CVEs block the release.
- Pre-existing HIGH/CRITICAL CVEs do NOT block but **MUST be filed as
tech-debt** before the release lands. File them in TWO TIERS (batching
rule, [org-internal #3846] — a single release audit once fanned out to 14 TDs → 14 PRs →
rule, [org-internal #3846] — a single release audit once fanned out to 14 TDs → 14 PRs →
56-84 CI runs on a saturated runner pool): Tier A below is the general
mechanical-change batching rule; Tier B and the escape hatch follow it.
@@ -86,7 +84,7 @@ Two source types today: (a) dependency bump audits; (b) docs 批量修订
- **(a) Dependency bump audits** (the [org-internal #3846] origin case): advisories whose
remediation is a plain version bump (lockfile-only diff, no semver-major
jump, no overrides/catalog surgery, no API or adapter fallout) are filed
together via `工单 API(见 TERMINOLOGY)create` with
together via REST `POST .../issues` with
`title="[{origin}] audit-batch@{version}: mechanical bumps for {N} advisories (TD-{NNN})"`
and labels `[tech_debt_label_id, severity_label_id, kind_bug_label_id]`
(severity = highest in the batch). One row per advisory — CVE/GHSA ID,
@@ -101,7 +99,7 @@ Two source types today: (a) dependency bump audits; (b) docs 批量修订
`title="[{origin}] docs-batch: {class} revisions for {N} files"`, with one
row per file (path, correction, reason); same ONE-batch-PR landing.
- **Landing**: when members live on separate branches, compose the single
batch PR via the `land-batch` skill (`core/skills/land-batch/SKILL.md`
batch PR via the `land-batch` skill (`.octopus/skills/land-batch/SKILL.md`
— cross-branch batch composition, topology B). For dependency-bump
batches pass `batch-compose --convergent bun.lock` (script
`.gitea/scripts/batch-compose.ts`): the regenerable lockfile is exempt
@@ -110,14 +108,14 @@ Two source types today: (a) dependency bump audits; (b) docs 批量修订
#### 适用边界 / Applicability boundary
- **非mechanical(含判断成分的变更,不并单)** — items requiring judgment
- **非 mechanical(含判断成分的变更,不并单)** — items requiring judgment
(wording decisions, behavior/API changes, review-dependent edits) are
never batched; file each as its own issue.
- **路径重叠(batch 成员触碰相同文件 → 不并单;唯一豁免 =
可再生共享文件`bun.lock` 经batch-compose `--convergent` 收敛)** — the sole
可再生共享文件 `bun.lock` 经 batch-compose `--convergent` 收敛)** — the sole
path-overlap exemption is the regenerable shared lockfile under
`--convergent`; every other same-file collision stays un-batched.
- **semver-major 跳变 / overrides·catalog 手术 / adapter·peer 联动(→
- **semver-major 跳变 / overrides·catalog 手术 / adapter·peer 联动(→
Tier B 独立单,维持 per-package)** — these stay per-package Tier B
issues. Tier B and the escape hatch (below) survive this generalization
unchanged.
@@ -157,12 +155,12 @@ template and `_shared/gitea-write-patterns.md` Pattern 3.
| Check | Status |
| ----------------- | ----------------------------------------------------- |
| Clean workspace | ✅/ ❌ |
| Clean workspace | ✅ / ❌ |
| Branch | {branch name} |
| Build | ✅/ ❌ |
| Typecheck + Lint | ✅/ ❌ |
| Tests | ✅/ ❌(N passed, M failed) |
| Dependency audit | ✅/ ⚠️ N known CVEs (pre-existing) / ❌N new CVEs |
| Build | ✅ / ❌ |
| Typecheck + Lint | ✅ / ❌ |
| Tests | ✅ / ❌ (N passed, M failed) |
| Dependency audit | ✅ / ⚠️ N known CVEs (pre-existing) / ❌ N new CVEs |
```
If any gate fails except known CVEs, stop and report what failed.
@@ -309,7 +307,7 @@ credentials in the environment.
- `NODE_AUTH_TOKEN` — only if publishing to npm.
- `AUR_KEY` — only if pushing the AUR package.
- Docker logged in to the Gitea container registry:
`echo "$GITEA_TOKEN" | docker login <instance-registry-host> -u <user> --password-stdin`.
`echo "$GITEA_TOKEN" | docker login eightarms.net -u <user> --password-stdin`.
### Steps
@@ -323,7 +321,7 @@ credentials in the environment.
```bash
OCTOPUS_VERSION={version} OCTOPUS_RELEASE={release-id} \
GH_REPO=Octopus/octopus GITEA_TOKEN=<pat> \
bun <harness-package>/script/build.ts
bun packages/octopus/script/build.ts
```
`build.ts` attaches the 12 release assets to the draft release.
@@ -339,14 +337,14 @@ credentials in the environment.
4. **Undraft the release** (makes it public):
```bash
curl -fsS -X PATCH \
"<instance-base-url>/api/v1/repos/Octopus/octopus/releases/{release-id}" \
"https://eightarms.net/api/v1/repos/Octopus/octopus/releases/{release-id}" \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/json" -d '{"draft": false}'
```
5. **Verify assets** — the release must carry 12 assets:
```bash
curl -sS "<instance-base-url>/api/v1/repos/Octopus/octopus/releases/tags/v{version}" \
curl -sS "https://eightarms.net/api/v1/repos/Octopus/octopus/releases/tags/v{version}" \
-H "Authorization: token $GITEA_TOKEN" | jq '.assets | length'
```
@@ -375,9 +373,14 @@ Document how to undo this release:
1. **Git rollback**:
- `git tag -d v{version}` (if not yet pushed).
- `git revert {commit_hash}` (if already merged).
- If the release includes DB migrations, confirm the `down` migration exists.
- If the release includes DB migrations, confirm the `down` migration exists
**when the project's migration tooling provides one**. This repo's Drizzle
convention is forward-only (`migration.sql` + `snapshot.json`, no `down`
files) — there, record the reversal plan (manual SQL / restore) or mark the
change irreversible instead of waiting on a `down` file that never comes.
2. **Data rollback** (if applicable):
- For each migration, confirm the `down` migration exists and has been tested.
- For each migration, confirm the `down` migration exists and has been tested
where the tooling offers one; otherwise document the manual reversal steps.
- If the release changes data format without a reversible migration, mark
`⚠️ IRREVERSIBLE DATA CHANGE`.
@@ -390,7 +393,7 @@ git tag -d v{version}
# or: git revert {hash}
### Data Rollback
- Migration `{name}`: down {exists / NOT FOUND}
- Migration `{name}`: down {exists / NOT FOUND / N/A — forward-only tooling}
- {additional risks}
### Rollback Triggers
@@ -419,7 +422,7 @@ git tag -d v{version}
**Version**: {old} → {new} (MAJOR / MINOR / PATCH)
**Tag**: v{version}
**Commits**: {N} since last tag ({feat} features, {fix} fixes)
**Typecheck**: ✅| Lint: ✅| Tests: ✅
**Typecheck**: ✅ | Lint: ✅ | Tests: ✅
### Changelog
@@ -435,13 +438,13 @@ git tag -d v{version}
**Deploy**: {manual step — "merge to main triggers CI", etc.}
**Rollback**: see Phase 5 above
**Next**: `core/skills/retrospective/SKILL.md` (optional — run on any project state)
**Next**: `.octopus/skills/retrospective/SKILL.md` (optional — run on any project state)
```
---
## References
- `core/checklists/release.md` — Release checklist
- `.octopus/checklists/release.md` — Release checklist
- Semver spec: https://semver.org
- Conventional Commits: https://www.conventionalcommits.org
+10 -1
View File
@@ -187,6 +187,15 @@ The audit target is the **SDLC factory itself** — the set of skills, checklist
- [ ] `core/skills/` exists and contains at least one `SKILL.md`.
- [ ] `core/checklists/audit-process.md` is accessible.
- [ ] `core/schemas/` contains the shared review schemas (`reviewer-output`, `synthesis`, `review-status`).
- [ ] **Process-tests-green gate ([org-internal #4787] STD 8.8)**: before dispatching
auditors, the Orchestrator (bash-capable; Explorer auditors have none by
construction) runs the process's own suite and guards — from
`<harness-package>`: `bun run test:parallel`, plus the repo-root `check:*`
guard family — and writes the commands, baseline SHA, and per-command result
summary to `<runs-root>/audit/reviews/audit-process/process-tests-green.md`
(Tier 1). The read-only STD 8.8 auditor verifies that record against the
current baseline; any red test/guard is itself a process defect and MUST be
filed as a finding.
No `.artifacts/{slug}/` inputs are required — the audit reads `<instance-root>/` itself. Audit outputs are published as Gitea wiki pages under `audit/{date}/`.
@@ -201,7 +210,7 @@ No `.artifacts/{slug}/` inputs are required — the audit reads `<instance-root>
| 5 | FLOW | Pipeline Cohesion | FLOW 5.1–5.6 | Preconditions + output sections of every `SKILL.md` |
| 6 | SCH | Schema Contract | SCH 6.1–6.5 | `core/schemas/*.json` + skills that reference them |
| 7 | DUP | Duplication & Drift | DUP 7.1–7.4 | ALL `SKILL.md` + ALL `checklists/*.md` |
| 8 | STD | Audit Standards (ISO/IEEE) | STD 8.1–8.7 | `review-*` skills, `review-artifact` (target: `audit-process`) itself |
| 8 | STD | Audit Standards (ISO/IEEE) | STD 8.1–8.8 | `review-*` skills, `review-artifact` (target: `audit-process`) itself |
| 9 | AGT | AGENTS.md Conformance | AGT 9.1–9.5 | ALL `SKILL.md` + `AGENTS.md` |
| 10 | TRC | Traceability | TRC 10.1–10.5 | `review-*` skills, schemas, status artifacts |
+1 -1
View File
@@ -860,7 +860,7 @@ After the Developer completes revision:
**Convergence criteria**: see `core/skills/_shared/review-pipeline-phases.md` Phase D for the canonical table (stop conditions). `{MAX_ROUNDS}` is per-skill — code review binds the routing budget: **2** rounds default, **3** for high risk (a BLOCKER survived a round, or the diff touches a frozen cross-session contract) — `<instance-root>/workflow-routing.yaml` `review.max_rounds` ([org-internal #3567]). The code-review-specific addition is inheritance of the shared INFO-disposition rule. The shared Round-3 Escalation Guardrail is scoped to the `review-artifact` skill targets only and does not apply here.
**Round budget (mandatory, [org-internal #3567])**: when `current_round` reaches the budget (2; 3 for high risk) without convergence, do NOT enter another round. STOP: register every remaining BLOCKER/MAJOR finding as TD registry rows (`rules/ticket-lifecycle.md`), record the verdict (FAIL if any BLOCKER remains, else PASS — the debt is carried by the TD registry rows just registered), and close the loop. A review that cannot converge within budget is a design-level stall wearing review clothes (retro 2026-08-21 / [org-internal #3050]): in DAG task mode, map it to the DAG oversize signals (`<instance-root>/workflow-routing.yaml` `dag.size_derivation.oversize_signals`) and let the aggregator agent drive re-derivation + a `review-dag` re-run; otherwise record the design-level finding on the Epic, or open `{slug}/03-design-amendments` via `wiki 读写 API(见 TERMINOLOGY)` (`_shared/gitea-write-patterns.md` Pattern 1).
**Round budget (mandatory, [org-internal #3567])**: when `current_round` reaches the budget (2; 3 for high risk) without convergence, do NOT enter another round. STOP: register every remaining BLOCKER/MAJOR finding as TD registry rows (`rules/ticket-lifecycle.md`), record the verdict with the shared three-tier ladder (`core/schemas/synthesis.schema.json` `overall_verdict` + `core/skills/_shared/review-orchestrator-rules.md` rule 3): FAIL if any BLOCKER or FAIL dimension remains, else WARN, PASS only if every remaining dimension is PASS ([org-internal #4786]: the prior binary "else PASS" wording silently upgraded WARN-eligible outcomes to PASS), and close the loop. The debt is carried by the TD registry rows just registered, not by the recorded verdict. A review that cannot converge within budget is a design-level stall wearing review clothes (retro 2026-08-21 / [org-internal #3050]): in DAG task mode, map it to the DAG oversize signals (`<instance-root>/workflow-routing.yaml` `dag.size_derivation.oversize_signals`) and let the aggregator agent drive re-derivation + a `review-dag` re-run; otherwise record the design-level finding on the Epic, or open `{slug}/03-design-amendments` via `wiki 读写 API(见 TERMINOLOGY)` (`_shared/gitea-write-patterns.md` Pattern 1).
**Efficiency optimization**: see `core/skills/_shared/review-pipeline-phases.md` Phase D (same rule — from round 2 onward ([org-internal #2598]), only spawn reviewers for FAIL/WARN/UNRESOLVED dimensions; full sweep every 2 rounds). Round ≥2 reviewer input is delta-scoped to the revised files + their neighbors.
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "octopus-workflow",
"version": "0.1.3",
"version": "0.1.4",
"description": "Standalone workflow core: provider-neutral agent workflow corpus (skills / rules / checklists / templates / schemas) with drift guards, extracted from the octopus agent platform.",
"license": "MIT",
"private": false,