Skip to content

Backfill Lessons Learned sections in 10 originating plan documents#326

Merged
clcollins merged 2 commits into
mainfrom
srepd/lessons-learned-backfill
Jun 10, 2026
Merged

Backfill Lessons Learned sections in 10 originating plan documents#326
clcollins merged 2 commits into
mainfrom
srepd/lessons-learned-backfill

Conversation

@clcollins

Copy link
Copy Markdown
Owner

Summary

  • Analyzed all 112 plan documents to identify fix relationships where a later plan corrected a flaw from an earlier one
  • Found 15 genuine fix relationships and added Lessons Learned sections to the 10 originating plan documents that were missing them
  • Each lesson follows the Agentic SDLC convention: category (GENUINE ERROR / PROCESS GAP), what happened, why it wasn't caught, what should change
  • Docs-only change — no code modifications

Highlights

Notable patterns discovered across the fix chain analysis:

  1. Narrow fixes missing other instances (009→013→059): fixing one handler but not auditing all handlers for the same bug class
  2. Incomplete deprecation (083→084): removing runtime behavior but leaving stale references in templates/help/README
  3. Hardcoded assumptions (055-claude→061): hardcoding external binary names that break in different environments
  4. Silent failures (003→058): CI integrations that fail without error, just missing data
  5. Dev/prod divergence (055-dev→066): dev fixtures that don't mirror production API structure
  6. 055-claude-cli-integration spawned 3 fix plans — the most prolific originator, suggesting the original design needed more upfront consideration of input isolation and dispatch extensibility

Test plan

  • Docs-only change, no code affected
  • grep -c "## Lessons Learned" docs/plans/*.md confirms 15 files now have the section
  • Each new section cross-references the fix plan doc by filename
  • CI passes (no code changes, only markdown)

🤖 Generated with Claude Code

clcollins and others added 2 commits June 10, 2026 08:32
Backfill Lessons Learned sections following the Agentic SDLC convention
(059-lessons-learned-convention). Analyzed all 112 plan docs to identify
15 genuine fix relationships where a later plan corrected a flaw from an
earlier one. Each lesson documents what happened, why it wasn't caught,
and what should change to prevent recurrence.

Co-Authored-By: Claude Opus 4.6 (1M context) <[email protected]>
Co-Authored-By: Claude Opus 4.6 (1M context) <[email protected]>
@codecov-commenter

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 67.60%. Comparing base (366d7ec) to head (5adcb05).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main     #326      +/-   ##
==========================================
- Coverage   68.30%   67.60%   -0.71%     
==========================================
  Files          46       49       +3     
  Lines        7323     7568     +245     
==========================================
+ Hits         5002     5116     +114     
- Misses       1988     2110     +122     
- Partials      333      342       +9     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@clcollins clcollins merged commit 10e8204 into main Jun 10, 2026
10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants