원클릭으로
manage-adr
// Manage Architecture Decision Records (ADRs). Use this to initialize, create, list, and link ADRs to document architectural evolution. Requires 'adr-tools' to be installed.
// Manage Architecture Decision Records (ADRs). Use this to initialize, create, list, and link ADRs to document architectural evolution. Requires 'adr-tools' to be installed.
Build the project and automatically fix packaging or build errors (for example Hatch failures) and related breakage. Use when the project fails to build, shows "broken" states, or after making significant changes.
Run CodeQL security/quality analysis and fix findings. Use when the user asks to run CodeQL, security scan, static analysis, or fix CodeQL findings.
Interactive deep research and decision support: frame the real problem (XY-aware), ask exactly 10 multiple-choice questions one at a time, then produce a rigorous comparative evaluation (default 5 approaches, 0–100 scores) and recommendation. Use when the user wants structured discovery before committing to a solution, a scored comparison of approaches, or to avoid jumping straight to an answer—especially for architecture, strategy, or high-stakes trade-offs.
Run linters and fix violations, formatting errors, or style mismatches using Trunk. Use when code quality checks fail, before submitting PRs, or to repair "broken" linting states.
At the end of a coding agent session (Cursor, Claude Code, Codex, Gemini CLI, or similar), summarize outcomes, failures, inefficiencies, and root causes, then output a concise postmortem with ranked Must/Should/Consider improvements. Chat-only output; do not edit project files unless the user explicitly asks. Skip nit-picks and one-off mistakes.
Broadly and deeply analyze user intent (avoiding XY problems) and evaluate multiple solution approaches (default 5) with scores from 0 to 100.
| name | manage-adr |
| description | Manage Architecture Decision Records (ADRs). Use this to initialize, create, list, and link ADRs to document architectural evolution. Requires 'adr-tools' to be installed. |
Architecture Decision Records (ADRs) are a lightweight way to document the "why" behind significant technical choices.
Use ADRs for decisions that meet one or more of the following criteria:
Do NOT use ADRs for:
A clear distinction must be maintained:
When an ADR requires implementation, link to design docs or relevant code in the References section.
Put these in pyproject.toml, Ruff/Pyright config, tests, or package README instead:
your_package.api”—not five relative paths.Granularity and anti-patterns: references/adr-granularity.md.
mend-adrmanage-adr: authoring and lifecycle (create, supersede, link, organize).mend-adr: drift workflow when behavior may no longer match an Accepted ADR—short, decision-level updates; keep volatile detail in code and config.Command reference: .claude/commands/mend-adr.md.
If ADRs are not yet initialized in the project, run:
adr init docs/adr
This ensures records are created in docs/adr.
To create a new ADR, use the provided script to ensure non-interactive creation:
.claude/skills/manage-adr/scripts/create-adr.sh "Title of the ADR"
After creation, the script will output the filename. You MUST then edit the file to fill in the Context, Decision, and Consequences.
If a new decision replaces an old one, use the -s flag:
.claude/skills/manage-adr/scripts/create-adr.sh -s <old-adr-number> "New Decision Title"
To link two existing ADRs (e.g., ADR 12 amends ADR 10):
adr link 12 Amends 10 "Amended by"
adr listread_file docs/adr/NNNN-title.mdadr generate tocadr generate graph | dot -Tpng -o adr-graph.pngPrefer conceptual diagrams (data flow, trust boundaries, user-visible behavior) over file-tree diagrams unless the ADR is about layout. Put diagrams in the Decision (or Architecture) section using a mermaid fenced block; add one short sentence of context before the block.
Examples and diagram types: references/mermaid-diagrams.md.
docs/adr/template.md.