Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
delphine-l avatar

Documentation

  • 29 installs
  • 17 repo stars
  • Updated May 14, 2026
  • delphine-l/claude_global

Best practices for session documentation during Claude Code work, including incremental fix summaries and audit trails.

About

Describes how to write incremental summary documents, fix reports, and audit trails during coding sessions. A developer uses it to keep a clear record of what was changed and why across multiple fixes.

  • Incremental summary docs after each major fix
  • Audit-trail structure for session changes

Documentation by the numbers

  • 29 all-time installs (skills.sh)
  • Ranked #939 of 1,879 Documentation skills by installs in the Skillselion catalog
  • Data as of Jul 29, 2026 (Skillselion catalog sync)
npx skills add https://github.com/delphine-l/claude_global --skill documentation

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs29
repo stars17
Last updatedMay 14, 2026
Repositorydelphine-l/claude_global

What it does

Best practices for session documentation during Claude Code work, including incremental fix summaries and audit trails.

Files

SKILL.mdMarkdownGitHub ↗

Documentation Skills

Best practices for creating effective documentation during Claude Code sessions.

Session Documentation

Incremental Summary Documents

When working through multiple related issues in a session:

1. Create summary after each major fix:

  • FIX_SUMMARY.md - First issue resolved
  • FISH_CLASSIFICATION_FIX_SUMMARY.md - Second issue
  • TAXONOMY_COMPLETE_SUMMARY.md - Final state

2. Benefits:

  • User can reference specific fixes
  • Clear audit trail of changes
  • Easy to see what was done and why
  • Supports future troubleshooting

3. Include in each summary:

  • Problem statement: What was the issue?
  • Root cause: Why did it happen?
  • Species/items affected: With counts for verification
  • Solution applied: Step-by-step what was done
  • Files modified: Complete list with paths
  • Verification results: Show the fix worked
  • Next steps: What to do with the results

Example Structure

# [Issue Name] - [STATUS]

## Problem
[Clear description of what was wrong]

## Root Cause
[Why the problem occurred]

## Items Fixed (N total)
| Item | Details | Category |
|------|---------|----------|
| ... | ... | ... |

## Solution Applied
1. [Step 1]
2. [Step 2]
...

## Verification Results
✓ [What was verified]
✓ [Counts match]

## Files Modified
1. `file1.txt` - [what changed]
2. `file2.txt` - [what changed]

## Next Steps
[What the user should do now]

When to Create Summaries

  • After fixing a significant bug: Document root cause and solution
  • After completing a major task: Show what was accomplished
  • After each iteration in multi-step fixes: Track progress
  • At session end: Comprehensive final state summary

Documentation File Naming

Use descriptive names that indicate:

  • What was fixed: FISH_CLASSIFICATION_FIX_SUMMARY.md
  • Scope: TAXONOMY_COMPLETE_SUMMARY.md
  • Status: Include "FIXED", "COMPLETE", "FINAL" in filename

Avoid generic names like SUMMARY.md or NOTES.md.

Related Skills

  • Skill Management: Capturing learnings from documentation
  • Token Efficiency: Concise yet complete documentation

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.