Adversarial design review of a System Design Doc. Interrogates the staff questions, finds the cost-explosion and failure risks, returns a verdict. Sensors and evals gate.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
Review a System Design Doc the way a staff engineer runs a design review. Follow guides/pipeline.md.
Ask once if missing: path to the design doc (default latest under
`.claude/runtime/outputs/architect/design/`). Reviews own or external designs.
Read:
- the target design doc
- guides/design-method.md (judge against the canon and the three pillars)
- guides/writing-style.md
- guides/templates/design-review.md
- guides/pipeline.md
Interrogate, do not summarize. For each of the 10 staff questions, give the design's answer and the
gap. Order findings by severity (blocker / high / medium / low). Name trade-offs the design left
implicit. Be specific: cite the section, name the failure, propose the fix. Default to skeptical.
Verdict: ship | revise | block.
Save to .claude/runtime/outputs/architect/review/{feature_id}.md (reuse the design's feature_id).
Sensors: sensors/design-structure.md (review variant: all 10 questions answered, verdict present).
Evals: evals/design-review-depth.md.
After save reply: Review saved at {path}. Verdict: {ship|revise|block}. Score: {N}/10.
The line at the top of this page pins the label git:20260615.091526b, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
A 17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
pass: Frontmatter block present
pass: Frontmatter declares a name
pass: Frontmatter declares a description
pass: Size between 200 bytes and 200 KB (1316 bytes)
pass: No zero-width or bidi control characters
pass: No instruction hidden inside an HTML comment
pass: No link to an exfiltration or paste host
pass: No credential-shaped string
pass: No instruction to send local credentials anywhere
pass: No text hidden with inline styles
pass: No prompt-injection phrasing
pass: No curl or wget piped into a shell
pass: No recursive delete of root, home or parent
pass: No instruction to read or print local credentials
pass: No base64 blob over 200 characters
pass: No link to a raw IP address
pass: No script tag
Source
GitHub
pierry/harness-kit · 4 stars · license MIT · pushed 2026-10-05 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_sfpqy4fylrwdkg3q
GET https://markdownregistry.com/api/v1/resolve?ref=pierry/harness-kit/review
GET https://markdownregistry.com/api/v1/blob/b263e2c8223675098ed0b19056e7319ccb2fb51470eb3a3823ce08e9d8cf0ca9
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
pierry/harness-kit · .claude/agents/product-manager/skills/prd/SKILL.md · Generate a Product Requirements Document for an team squad. Business-facing artifact. Sensors and evals gate.
pierry/harness-kit · .claude/agents/product-manager/skills/prp/SKILL.md · Generate a Product Requirements Prompt for engineering handoff. Needs an approved PRD. Sensors, link validation, and…
pierry/harness-kit · .claude/agents/staff-software-engineer/skills/backend/SKILL.md · Backend engineering conventions for team. Java/Kotlin servers. Project conventions in .claude/conventions/backend.md…
pierry/harness-kit · .claude/agents/staff-software-engineer/skills/mobile/SKILL.md · Mobile engineering conventions for team. iOS and Android. Project conventions in .claude/conventions/mobile.md override…
pierry/harness-kit · .claude/agents/staff-software-engineer/skills/web/SKILL.md · Web frontend engineering conventions for team. Vue.js stack. Project conventions in .claude/conventions/web.md override…
pierry/harness-kit · .claude/agents/system-architect/skills/design/SKILL.md · Design any system from scratch into a rigorous System Design Doc. Generic fallback when no topic-specific playbook…
pierry/harness-kit · .claude/agents/system-architect/skills/search-engine/SKILL.md · Design a web-scale search engine (crawler, processing, indexing, query serving). Topic playbook from the System Design…
qwenlm/qwen-code · packages/core/src/skills/bundled/review/SKILL.md · Review changed code for correctness, security, code quality, and performance. Use when the user asks to review code…
suboss87/fdeops · skills/review/SKILL.md · Review a proposed customer code change against its intended outcome and operational risks. Use for a diff or PR review…
snowflake-labs/cocoplus · .cortex/skills/cocobrew/review/SKILL.md · Enter the Review phase of CocoBrew. Aggregates findings from Code Quality Advisor, CocoCupper intelligence, spec…
