Run a full Section 508 + WCAG 2.2 Level AA accessibility audit on a federal / .gov / public-sector website or ICT product. Triggers: "Section 508", "508 audit", "WCAG", "VPAT", "ACR".
日本語の概要は準備中です。原文の説明を表示しています。
Create and manage Architecture Decision Records using the Michael Nygard format. Triggers: you need to document architecture decisions, record technical choices, explain why a technology was chosen.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
You are in AUTONOMOUS MODE. Do NOT ask questions. Create or manage Architecture Decision Records for this project.
INPUT: $ARGUMENTS
Accepted arguments:
retrospective or retro: Analyze the codebase and generate ADRs for decisions
already made but not documented.index: Regenerate the ADR index file.supersede <N> <new decision>: Create a new ADR that supersedes ADR-N.deprecate <N>: Mark ADR-N as deprecated.list: List all existing ADRs with their statuses.If no arguments provided, run in retrospective mode.
Step 1.1 -- Initialize ADR Directory
Check if docs/adr/ exists. If not, create it.
Check for ADRs in alternative locations:
adr/docs/decisions/decisions/doc/adr/If found in an alternative location, use that location instead of creating a new one.
Step 1.2 -- Determine Next ADR Number
Scan existing ADR files matching the pattern NNNN-*.md (e.g., 0001-use-postgresql.md).
Find the highest number and increment by 1 for the next ADR.
If no ADRs exist, start at 0001.
Step 1.3 -- Read Existing ADRs
Read all existing ADR files to understand:
When creating a new ADR from a decision description:
Step 2.1 -- Research the Decision
Analyze the codebase to gather context for the decision:
Step 2.2 -- Write ADR
Use the Michael Nygard format:
# [ADR-NNNN] [Title: Short Decision Statement]
**Date:** YYYY-MM-DD
**Status:** [proposed | accepted | deprecated | superseded by [ADR-XXXX](XXXX-title.md)]
## Context
[What is the issue that we're seeing that is motivating this decision or change?
Describe the forces at play: technical, political, social, project-specific.
These forces are likely in tension and should be called out as such.
State facts, not opinions. 2-4 paragraphs.]
## Decision
[What is the change that we're proposing and/or doing?
State the decision clearly and unambiguously.
Use active voice: "We will use X" not "X should be used."
1-2 paragraphs.]
## Consequences
[What becomes easier or more difficult to do because of this change?
Include both positive and negative consequences.
Be honest about trade-offs.
**Positive:**
- [consequence]
- [consequence]
**Negative:**
- [consequence]
- [consequence]
**Neutral:**
- [consequence]
]
Step 2.3 -- File Naming
Name the file: NNNN-kebab-case-title.md
0003-use-postgresql-for-primary-datastore.mdWhen running in retrospective mode, analyze the codebase to infer architectural decisions that were made but never documented:
Step 3.1 -- Detect Decisions
Scan for these categories of decisions:
| Category | What to Look For |
|---|---|
| Language/Runtime | Primary language, runtime version constraints |
| Framework | Web framework, UI framework, CLI framework choice |
| Database | Database type, ORM/query builder choice |
| Auth | Authentication strategy, provider choice |
| State Management | State management library/pattern |
| API Style | REST vs GraphQL vs gRPC, API versioning strategy |
| Infrastructure | Docker, K8s, serverless, cloud provider |
| CI/CD | Pipeline tool, deployment strategy |
| Testing | Test framework, testing strategy (unit/integration/e2e) |
| Monorepo vs Polyrepo | Repository structure decision |
| Dependency Management | Package manager, lockfile strategy |
| Code Style | Linter, formatter, style guide |
| Error Handling | Error handling strategy, logging approach |
| Caching | Caching strategy, cache provider |
Step 3.2 -- Generate ADRs
For each detected decision:
Step 3.3 -- Prioritize
Generate retrospective ADRs in priority order:
Generate or update docs/adr/README.md (or docs/adr/index.md):
# Architecture Decision Records
This directory contains Architecture Decision Records (ADRs) for [Project Name].
## What is an ADR?
An ADR is a short document that captures an important architectural decision
made along with its context and consequences.
## ADR Index
| # | Title | Status | Date |
|---|-------|--------|------|
| [0001](0001-title.md) | Title | Accepted | YYYY-MM-DD |
| [0002](0002-title.md) | Title | Superseded by [0003](0003-title.md) | YYYY-MM-DD |
| ... | ... | ... | ... |
## Statuses
- **Proposed** -- Under discussion, not yet decided
- **Accepted** -- Decision has been made and is in effect
- **Deprecated** -- Decision is no longer relevant
- **Superseded** -- Replaced by a newer decision (linked)
After producing documentation, validate completeness:
IF VALIDATION FAILS:
| # | Title | Status | Date |
|---|---|---|---|
| NNNN | Title | Status | Date |
| ... | ... | ... | ... |
docs/adr/NNNN-title.md -- [description]docs/adr/README.md -- Updated indexAfter producing output, record execution metadata for the /evolve pipeline.
Check if a project memory directory exists:
~/.claude/projects/skill-telemetry.md in that memory directoryEntry format:
### /adr — {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Self-healed: {{yes — what was healed | no}}
- Iterations used: {{N}} / {{N max}}
- Bottleneck: {{phase that struggled or "none"}}
- Suggestion: {{one-line improvement idea for /evolve, or "none"}}
Only log if the memory directory exists. Skip silently if not found. Keep entries concise — /evolve will parse these for skill improvement signals.
NEXT STEPS:
After creating ADRs:
/document to check overall documentation health."/diagram to generate architecture diagrams that complement these ADRs."/arch-review to review the architecture decisions for consistency."まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Run a full Section 508 + WCAG 2.2 Level AA accessibility audit on a federal / .gov / public-sector website or ICT product. Triggers: "Section 508", "508 audit", "WCAG", "VPAT", "ACR".
日本語の概要は準備中です。原文の説明を表示しています。
Analyzes actuarial modeling systems for loss reserving accuracy, premium pricing methodology, mortality/morbidity tables, stochastic modeling, and capital adequacy per SOA and Solvency II standards..
日本語の概要は準備中です。原文の説明を表示しています。
Analyzes advertising yield across programmatic and direct-sold inventory including CPM/CPC/CPA performance, header bidding efficiency, fill rate optimization, inventory monetization strategies, and compliance with IAB standards and OpenRTB protocols..
日本語の概要は準備中です。原文の説明を表示しています。
Analyzes affordable housing management software for unit allocation algorithms, waitlist management, income verification workflows, Fair Housing compliance, LIHTC and Section 8 program tracking, inspection scheduling, and tenant reporting..
日本語の概要は準備中です。原文の説明を表示しています。
Rewrites existing landing pages, blog posts, and documentation so AI search engines (Google AI Overviews, ChatGPT, Perplexity, Claude, Gemini) cite them. Triggers: an existing page underperforms in AI answers despite ranking well in classic SERPs.
日本語の概要は準備中です。原文の説明を表示しています。
Track brand mentions, URL citations, and share-of-voice across the 2026 AI search surface — ChatGPT (with browsing), Perplexity, Claude (with search), Google Gemini, Google AI Overviews, Bing Copilot, You.com, Phind, and Microsoft Copilot..
日本語の概要は準備中です。原文の説明を表示しています。