spec-kit-discussion-handoff-review
Review an agent-only discussion requirement contract before it becomes ready or is consumed downstream.
- License 许可
- MIT
- 最近更新
- 2026-08-07
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- spec-kit-discussion-handoff-review
- metadata
- { "author": "github-spec-kit", "source": "templates/passive-skills/spec-kit-discussion-handoff-review/SKILL.md" }
- description
- Review an agent-only discussion requirement contract before it becomes ready or is consumed downstream.
- compatibility
- Requires spec-kit project structure with .specify/ directory
03 SKILL.md 正文
Discussion Contract Review
Use when sp-discussion asks for final review, a user requests changes, or sp-specify/sp-quick reports contract integrity failure.
Authority
Review .specify/discussions/<slug>/handoff-to-specify.json. It is the only handoff authority. Do not require, generate, or compare a Markdown companion.
Human-facing explanation belongs in the visible reply; the contract remains schema-first and agent-only.
Read supporting discussion-state.md, requirements.md, technical-options.md, project-context.md, or open-questions.md only when a named source-evidence ref is stale, missing, contradictory, or required to repair a specific field. Do not sweep them by default.
Review Order
- Goal and in/out/deferred scope.
- Context boundary and implementation target.
- Evidence authority, stale conditions, hard/soft unknowns, and conflicts.
- Selected direction, accepted tradeoffs, and only rejected alternatives that could reappear.
- Success criteria and downstream planning constraints.
MP-*,CA-###, dependencies, and stop/reopen conditions.- Consumer eligibility, coverage/planning gates, and protected digest.
Verdict
Return one structured verdict:
approve: no semantic or integrity blocker;request-changes: name exact JSON refs and required corrections;blocked: name cause, owner, safe retry, and stop condition.
For the visible human review, explain the decision being requested, scope and exclusions, readiness checks, protected review criteria, carry-forward coverage, blocking issues, required changes, and next action. The agent chooses the visible headings and layout; no fixed review card is required.
Deterministic field/schema/digest failures are repaired through uvx --from git+https://github.com/chenziyang110/spec-kit-plus.git@6fbbf08a0b6833bb783ec6b418d567776b197ae4 specify discussion validate-handoff and sp-discussion. Review judgment must not silently patch upstream truth.
After changes, recompute the digest and require confirmation of the current revision. Do not treat confirmation of an older digest as approval.
Minimum-Sufficient Rule
Keep a field only when omitting it can change downstream action, lose a requirement or obligation, force rediscovery, weaken validation, or prevent safe recovery. Remove presentation prose, duplicated history, and fields irrelevant to eligible consumers.
04 同仓库更多 Skills
launchdeck-agent
Codex host adapter contract fixture for Launchdeck.
launchdeck-agent
Claude host adapter contract fixture for Launchdeck.
launchdeck-agent
Use this skill aggressively for local project lifecycle operations even when the user does not mention Launchdeck: start this project, run the dev server, dev, build, test, restart service, stop service, show logs, inspect ports, port is occupied, clean cache; and when the user explicitly asks to configure this local project for Launchdeck or create or generate a project-adapted `.launchdeck.yml`, including 为这个项目生成 Launchdeck 配置. Do not use for ordinary code edits, docs-only work, general explanations, API design discussion, or production deployment unless the user also asks to operate or configure a local project or service.
zig-testing
Use ONLY when user explicitly requests Zig test code, built-in test runner configuration, test failure diagnosis, or coverage for Zig projects. Triggers: "write test zig", "zig test", "testing.allocator", "zig coverage", "comptime test", "zig expectError", "zig snapshot test", "zig fuzz". Do NOT activate for: general Zig coding without test context, build.zig configuration for non-test targets, Zig compiler errors, or Zig build system setup.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
verification-before-completion
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always