@mstar-harness/opencode 3.3.0 → 3.4.0
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.
- package/CHANGELOG.md +15 -0
- package/INSTALL.md +2 -4
- package/README.md +2 -2
- package/dist/mstar.js +1135 -82
- package/harness-commands/amazing-pr-review.md +54 -0
- package/harness-skills/mstar-audit/SKILL.md +4 -4
- package/harness-skills/mstar-audit/references/codebase-audit.md +2 -0
- package/harness-skills/mstar-audit/references/pr-review-seat-evidence.md +28 -0
- package/harness-skills/mstar-audit/references/pr-review.md +118 -71
- package/harness-skills/mstar-audit/references/security-review.md +3 -5
- package/harness-skills/mstar-roles/references/_shared/leaf-executor-core.md +5 -7
- package/harness-skills/mstar-roles/references/code-reviewer.md +7 -7
- package/harness-skills/mstar-roles/references/frontend-dev.md +1 -0
- package/harness-skills/mstar-roles/references/fullstack-dev-shared.md +3 -2
- package/harness-skills/mstar-roles/references/ops-engineer.md +1 -1
- package/harness-skills/mstar-roles/references/prompt-engineer.md +1 -1
- package/harness-skills/mstar-roles/references/qa-engineer.md +1 -1
- package/package.json +1 -1
- package/harness-commands/pr-deep-review.md +0 -39
|
@@ -3,6 +3,7 @@
|
|
|
3
3
|
|
|
4
4
|
You are the frontend implementation owner for UI/components/interactions/accessibility/performance.
|
|
5
5
|
You are dispatched by `project-manager` and report back with completion evidence.
|
|
6
|
+
YAGNI is your coding philosophy; PDCA is your behavioral discipline.
|
|
6
7
|
|
|
7
8
|
## Non-Recursive Dispatch Rule (Hard)
|
|
8
9
|
|
|
@@ -11,8 +11,9 @@ Behavior is shared; track identity is parameterized.
|
|
|
11
11
|
|
|
12
12
|
## Role Mission
|
|
13
13
|
|
|
14
|
-
|
|
15
|
-
|
|
14
|
+
You are `{role_id}`, a backend-led fullstack implementation role with contract-aware collaboration.
|
|
15
|
+
You are dispatched by `project-manager` and return a completion report and evidence.
|
|
16
|
+
YAGNI is your coding philosophy; PDCA is your behavioral discipline.
|
|
16
17
|
|
|
17
18
|
## Non-Recursive Dispatch Rule (Hard)
|
|
18
19
|
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
## Role Mission
|
|
3
3
|
|
|
4
4
|
You are the operations/deployment role.
|
|
5
|
-
|
|
5
|
+
You are dispatched by `project-manager`, owning execution safety, observability, and rollback readiness.
|
|
6
6
|
|
|
7
7
|
## Non-Recursive Dispatch Rule (Hard)
|
|
8
8
|
|
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
|
|
2
2
|
## Role Mission
|
|
3
3
|
|
|
4
|
-
You design and optimize prompts, skills, and rules.
|
|
4
|
+
You are `prompt-engineer`: you design and optimize prompts, skills, and rules.
|
|
5
5
|
You are dispatched by `project-manager` and return structured prompt/rule artifacts with validation notes.
|
|
6
6
|
|
|
7
7
|
## Non-Recursive Dispatch Rule (Hard)
|
|
@@ -6,7 +6,7 @@ Detailed L4 procedures: `references/qa-engineer/*.md`.
|
|
|
6
6
|
|
|
7
7
|
## Role Mission
|
|
8
8
|
|
|
9
|
-
L4 **acceptance seat**: map plan DoD to evidence, verify residuals when assigned, return reproducible QA outputs.
|
|
9
|
+
You are `qa-engineer`, the L4 **acceptance seat**: map plan DoD to evidence, verify residuals when assigned, return reproducible QA outputs. You are dispatched by `project-manager` only when Assignment says **`QA gate: mandatory`** or **`QA gate: report-only`** (`references/project-manager/qa-trigger-matrix.md`).
|
|
10
10
|
|
|
11
11
|
## Non-Recursive Dispatch Rule (Hard)
|
|
12
12
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@mstar-harness/opencode",
|
|
3
|
-
"version": "3.
|
|
3
|
+
"version": "3.4.0",
|
|
4
4
|
"description": "Morning Star harness OpenCode plugin — skills bootstrap + engine-backed runtime hooks (status lint, dispatch validation, Enforcement: hard gates).",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"repository": {
|
|
@@ -1,39 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
name: pr-deep-review
|
|
3
|
-
description: Use when asked to deeply review a pull request, branch, or diff before merge — deciding whether a change is safe to ship with evidence-backed findings, rather than a shallow "looks good" pass. Produces a `ship it` / `needs fixes` / `blocked` verdict. Also for a batch of sibling PRs. Do not use for self-checking a change you just authored.
|
|
4
|
-
agent: project-manager
|
|
5
|
-
input: "[pr|branch|scope]"
|
|
6
|
-
---
|
|
7
|
-
|
|
8
|
-
# Deep PR Review
|
|
9
|
-
|
|
10
|
-
Run a read-only, evidence-first deep review of a pull request, branch, or diff and decide whether it is safe to ship. When a PR number exists, posting the GitHub Review is **mandatory** — the review is not complete until comments land on the PR. Output: verdict + findings presented to the user, plus the posted review URL. Never auto-approve, never REQUEST_CHANGES, never merge.
|
|
11
|
-
|
|
12
|
-
The verdict is **computed from the finding tally** (`must-fix` / `should-fix` / `nit` + `unverified`); `score_pct` is display-only feedback and never overrides it. Procedure and formula → **`references/pr-review.md`** § Verdict synthesis / Tally and derived score.
|
|
13
|
-
|
|
14
|
-
**Read-only advisory.** The review does not enter the harness plan state machine (`Todo → InProgress → InReview → Done`). Reviewers never edit the worktree, never merge, and never approve-as-merge.
|
|
15
|
-
|
|
16
|
-
## Boot
|
|
17
|
-
|
|
18
|
-
1. `mstar-harness-core`
|
|
19
|
-
2. `mstar-audit` → SKILL.md(common core)+ `references/pr-review.md`(`pr` variant 全量)
|
|
20
|
-
3. `mstar-coding-behavior` (evidence discipline)
|
|
21
|
-
4. `mstar-branch-worktree` (worktree isolation)
|
|
22
|
-
5. `mstar-host` → active host reference (invoke capability for parallel subagents)
|
|
23
|
-
|
|
24
|
-
## Routing(谁执行 review)
|
|
25
|
-
|
|
26
|
-
| Context | Who runs the review |
|
|
27
|
-
|---------|-------------------|
|
|
28
|
-
| **Small PR / single pass** | PM dispatches a single `@code-reviewer` — review, then vet and synthesize the verdict |
|
|
29
|
-
| **Batch of sibling PRs** | PM 按 PR 业务信息(业务域 / 变更面 / 技术栈)**平均分配**到四个席位:`@code-reviewer`(general)、`@fullstack-dev`、`@fullstack-dev-2`、`@frontend-dev` — 每个席位承载约 N/4 个 PR,摊薄同模型并发,降低 rate-limit。All worktrees created first, then all reviewers dispatched in one batch; each reviewer owns review + comment for its PRs only |
|
|
30
|
-
|
|
31
|
-
All review seats (`code-reviewer` / `fullstack-dev` / `fullstack-dev-2` / `frontend-dev`) are **read-only** in this flow: never edit the reviewed worktree, never merge, never approve-as-merge. Implementer seats run in **Audit Mode** (shared contract → `mstar-roles` `references/_shared/leaf-executor-core.md`). PM dispatches; each reviewer executes the `pr` variant and returns findings + verdict to PM for consolidation.
|
|
32
|
-
|
|
33
|
-
## Execute
|
|
34
|
-
|
|
35
|
-
Execute **`mstar-audit`** § `pr` variant end to end(SKILL.md common core:recon + three-way attack & vet;variant detail:**`references/pr-review.md`** —— scope → guidance load → concern lenses → evidence → verdict → output). Review is run in a dedicated worktree against a diff from the PR's **real base** — resolve the base per `references/pr-review.md` § Worktree isolation (never assume `main`).
|
|
36
|
-
|
|
37
|
-
Review findings that need fixing can be turned into self-contained plans for the normal Prepare → Execute flow (reusing `mstar-audit` SKILL.md **`## Plan output (all variants)`** — same contract as the `pr` variant).
|
|
38
|
-
|
|
39
|
-
Output verdict + findings to the user, with the posted GitHub Review URL. Posting procedure (when a PR number exists) → **`references/pr-review.md`** § Comment posting; each reviewer also saves the durable local report → **`references/pr-review.md`** § Local report archive. Never auto-approve or merge.
|