Implement a complete spec from its ticket graph, using isolated workers and one integration branch.
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
# Implement Spec
Build a complete specification from its dependency-aware ticket graph. The goal is a verified **integration branch**, not a collection of disconnected worker branches.
## When to Use
Use when:
- A reviewed spec exists.
- Tickets describe acceptance criteria and blocking relationships.
- The user wants the whole spec implemented rather than one ticket.
Do not use when the spec is still ambiguous, tickets are missing dependencies, or the user only wants one isolated ticket. Resolve those inputs first with the relevant planning skills.
## Contract
Before changing code, identify:
- The spec path or authoritative issue/plan.
- The ticket source and ticket status format.
- The dependency field used to define blocking relationships.
- The repository's test, lint, typecheck, and build commands.
- The base branch and the requested integration branch name.
If any of these cannot be found, stop and report the missing input instead of inventing a task graph.
## Workflow
### 1. Read the graph once
Read the full spec and all tickets. Extract a compact work map:
```text
Ticket Blocks on Files/area Acceptance check Status
…
The line at the top of this page pins the label v1.0.0, 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 (6605 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
jellydn/my-ai-tools · 123 stars · license MIT · pushed 2026-10-19 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_nqf2vbmsla57s4uf
GET https://markdownregistry.com/api/v1/resolve?ref=jellydn/my-ai-tools/implement-spec
GET https://markdownregistry.com/api/v1/blob/d03c90c86fb3dc404a888e205f13bd92c70298e08fe03b28e204fa7ad10f4669
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.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/accountable-engineering/SKILL.md · Use explicit human checkpoints for architecture, security, rollout, or operational decisions.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/babysit-pr/SKILL.md · Use when monitoring an open GitHub PR for CI failures, review feedback, mergeability, and safe retries or fixes.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/blindspot-pass/SKILL.md · Inspect history and architecture for hidden constraints before risky cross-cutting changes.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/capability-experiments/SKILL.md · Build an interactive report or experiment when the user asks to explore model capabilities.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/code-quality-review/SKILL.md · Audit a diff for structural maintainability and unnecessary complexity.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/code-review/SKILL.md · Review a branch, PR, or worktree diff for repository conventions and stated intent.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/codemap/SKILL.md · Create a seven-document architecture map when the user asks to map a codebase.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/commit-atomic/SKILL.md · Group staged changes into atomic commits with commitizen convention
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/context-discovery/SKILL.md · Use MCP tools to trace behavior that spans multiple modules or history sources.
dzhng/skills · skills/engineering/implement-spec/SKILL.md · Implement an existing spec through committed passes. Use for long or multi-pass specs that need maintenance checkpoints…
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/implement-spec/SKILL.md · Implement a complete spec from its ticket graph, using isolated workers and one integration branch.
dzhng/duet-agent · .agents/skills/implement-spec/SKILL.md · Implement an existing spec through committed passes. Use for long or multi-pass specs that need maintenance checkpoints…
leek/agent-skills · skills/implement-spec/SKILL.md · Build a whole spec fast: parallel implementer subagents work the ticket frontier, each in its own worktree, merging…
