Trellis Review Twostage
Before marking a Trellis subtask complete after self-check all green, conduct two-stage code review: mechanically check spec compliance (against PRD acceptance criteria/decision table/forbidden/file manifest/mount points), then code quality review (against .trellis/spec/guides engineering standards and design discipline). Used in Codex / Claude Code team collaboration scenarios, especially when implementation is by a small-parameter model like qwen3.6 35b—small models reviewing their own code is unreliable, Stage 2 should escalate to strong model; critical issues block completion.
Install
npx skills add coldwateryi/trellis-skills --skill trellis-review-twostagegit clone --depth 1 https://github.com/coldwateryi/trellis-skills.git /tmp/trellis-skills
mkdir -p ~/.claude/skills
cp -r /tmp/trellis-skills/trellis-review-twostage ~/.claude/skills/trellis-review-twostagePersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Trellis Two-Stage Review ## Overview Quality gate before marking subtask complete, after implementation self-check all green. Split into two independent review stages: - **Stage 1 Spec Compliance**: Mechanically check against `prd.md`, can be done by small model + checklist. - **Stage 2 Code Quality**: Check against engineering standards and design discipline, **should be done by strong model** (small model reviewing its own output is unreliable). Any stage finding **critical** issue blocks completion, send back to `trellis-implement-tdd` for fix; after fix, review again. Trigger: `trellis-implement-tdd` wrap-up self-check all green, handoff to review. Write every review pass to the fixed task-local file `<task-dir>/review-report.md`. Create it from `references/review-report-template.md` on the first review pass, then append a new review-round block for each re-review. Do not edit the template under the skill directory. ## Constraints - **Review based on diff + `prd.md`**, not re-reading entire requirements document. - **Stage 1 and Stage 2 separate, don't merge**—merging causes "did it follow spec" and "is code good" to mask each other. - **Stage 2 use strong model** (role-layered model assignment). If only small model available, Stage 2 at least strictly walk `references/review-stage2-checklist.md` item by item, and flag "not reviewed by strong model" risk in report.
Details
- Skill name
trellis-review-twostage- Repository
- coldwateryi/trellis-skills
- Path
trellis-review-twostage/SKILL.md- License
- MIT
- Repo updated
- 3 July 2026
FAQ
What does the Trellis Review Twostage skill do?
Before marking a Trellis subtask complete after self-check all green, conduct two-stage code review: mechanically check spec compliance (against PRD acceptance criteria/decision table/forbidden/file manifest/mount points), then code quality review (against .trellis/spec/guides engineering standards and design discipline). Used in Codex / Claude Code team collaboration scenarios, especially when implementation is by a small-parameter model like qwen3.6 35b—small models reviewing their own code is unreliable, Stage 2 should escalate to strong model; critical issues block completion.
How do I install the Trellis Review Twostage skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add coldwateryi/trellis-skills --skill trellis-review-twostage. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Trellis Review Twostage popular?
It lives in coldwateryi/trellis-skills, which has 44 GitHub stars and 1 forks.
Related skills
Trellis Debug Systematic
44coldwateryi/trellis-skills
When a test should be green but stays red, or a self-check command fails during Trellis subtask implementation/verification, use this non-deviable rigid script for systematic root-cause pinpointing and minimal fix. Used when Codex / Claude Code drives a small-parameter model like qwen3.6 35b—small models quickly spiral out of control when debugging freely; this skill uses "one change at a time, must re-run after change, no guessing" iron rules to lock it on a narrow path with defensive regression checks.
Trellis Debug Systematic Zh
44coldwateryi/trellis-skills
当 Trellis 子任务实现/验证过程中测试该绿不绿、或自检命令失败时,用一段不可偏离的刚性脚本做系统化根因定位与最小修复。用于 Codex / Claude Code 等驱动 qwen3.6 35b 这类小参数模型调试时——小模型一旦自由发挥就会到处乱改,本技能用"一次只改一处、改完必重跑、禁止猜"的铁律把它锁在窄路径上,并带防御性回归检查。
Trellis Delivery Controller
44coldwateryi/trellis-skills
Trellis requirements-to-delivery entrypoint and workflow controller. Use when the user wants to invoke one Trellis skill that detects the current stage and routes work across requirements analysis, MVP task planning, post-MVP gap audit, delivery loop, TDD implementation, systematic debugging, two-stage review, and final acceptance. Coordinates trellis-zero-to-mvp, trellis-mvp-to-delivery, trellis-implement-tdd, trellis-debug-systematic, and trellis-review-twostage while enforcing safety gates.
Agent Output Verifier
43rohitg00/agentbrain
Use when agent-produced work needs a safety, evidence, and reliability check before handoff, merge, or trust.
Snapshot
33sanghun0724/cmux-claude-skills
현재 cmux 레이아웃과 Claude 세션 ID를 JSON 스냅샷으로 저장합니다. cmux 재시작 시 복원에 사용됩니다.
Brain Eval
43rohitg00/agentbrain
Agent Brain /brain-eval: changing Agent Brain behavior or checking quality.
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.