Adr Code Traceability

ADR Code Traceability

Add Architecture Decision Record references to code for decision traceability. Provides language-specific patterns and placement guidelines.

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

When to Use This Skill

• Creating new files as part of an ADR implementation
• Documenting non-obvious implementation choices
• User mentions "ADR traceability", "code reference", "document decision"
• Adding decision context to code during /itp:go Phase 1

Quick Reference

Reference Format

ADR: {adr-id}

Path Derivation: ADR: 2025-12-01-my-feature → /docs/adr/2025-12-01-my-feature.md

Language Patterns (Summary)

See Language Patterns for complete examples.

Placement Decision Tree

Is this a NEW file created by the ADR?
├── Yes → Add reference in file header
└── No → Is the change non-obvious?
    ├── Yes → Add inline comment with reason
    └── No → Skip ADR reference

See Placement Guidelines for detailed guidance.

Examples

New File (Python)

"""
Redis cache adapter for session management.

ADR: 2025-12-01-redis-session-cache
"""

class RedisSessionCache:
    ...

Inline Comment (TypeScript)

// ADR: 2025-12-01-rate-limiting - Using token bucket over sliding window
// for better burst handling in our use case
const rateLimiter = new TokenBucketLimiter({ rate: 100, burst: 20 });

Do NOT Add References For

• Every line touched (only where traceability adds value)
• Trivial changes (formatting, typo fixes)
• Standard patterns (well-known idioms)
• Test files (unless test approach is an ADR decision)

Reference Documentation

• Language Patterns - Python, TS, Rust, Go patterns
• Placement Guidelines - When and where to add

Troubleshooting

Post-Execution Reflection

After this skill completes, check before closing:

1. Did the command succeed? — If not, fix the instruction or error table that caused the failure.
2. Did parameters or output change? — If the underlying tool's interface drifted, update Usage examples and Parameters table to match.
3. Was a workaround needed? — If you had to improvise (different flags, extra steps), update this SKILL.md so the next invocation doesn't need the same workaround.

Only update if the issue is real and reproducible — not speculative.