Step Name Alignment Issues
Found in: multiple .lock.yml files
Summary
Several workflows use Cache <X> as a step name for cache-save operations. The established project pattern (as used for Save qmd models to cache, Save qmd index to cache, Save cache-memory to cache (default)) is Save <X> to cache. The Cache <X> naming is ambiguous — it doesn't clarify whether data is being saved or restored — and is inconsistent with the rest of the codebase.
Issues Identified
[High Priority] Inconsistent pattern: "Cache (X)" → "Save (X) to cache"
Affected step names and files:
| Current Name |
Suggested Name |
Files Affected |
Cache node-llama-cpp binaries |
Save node-llama-cpp binaries to cache |
daily-doc-healer, daily-doc-updater, dev, developer-docs-consolidator, dictation-prompt, glossary-maintainer, smoke-codex, technical-doc-writer, unbloat-docs, weekly-blog-post-writer (10 files) |
Cache layout spec data |
Save layout spec data to cache |
layout-spec-maintainer (1 file) |
Cache prompt clustering data |
Save prompt clustering data to cache |
prompt-clustering-analysis (1 file) |
Cache downloaded data |
Save downloaded data |
daily-news (1 file — uses repo-memory, not actions/cache) |
Cache cache-memory file share data |
Restore cache-memory file share data |
ai-moderator (1 file — this step restores data, not saves it) |
Issue:
All of these steps either use actions/cache@... to save data to the GitHub Actions cache, or perform a custom save to repo-memory. The established naming pattern (Save <X> to cache) makes the save direction explicit. The ambiguous Cache <X> prefix is used correctly for save operations in qmd-related steps elsewhere, and these should match.
Note: Cache cache-memory file share data in ai-moderator.lock.yml is particularly confusing because it is actually a restore operation, not a save.
Established pattern references:
Save qmd models to cache ✅Save qmd index to cache ✅Save cache-memory to cache (default) ✅
Agentic Task Description
To fix these step names:
- Update source workflows — Change step names in the
.md source files (not the .lock.yml files directly):
Cache node-llama-cpp binaries → Save node-llama-cpp binaries to cache (in all 10 workflows)Cache layout spec data → Save layout spec data to cache (in layout-spec-maintainer.md)Cache prompt clustering data → Save prompt clustering data to cache (in prompt-clustering-analysis.md)Cache downloaded data → Save downloaded data (in daily-news.md)Cache cache-memory file share data → Restore cache-memory file share data (in ai-moderator.md)
- Recompile — Run
gh aw compile <workflow>.md for each changed source file
- Verify — Confirm updated
.lock.yml files reflect the new names
Related Files
- Compiled workflows:
.github/workflows/<name>.lock.yml
- Naming patterns:
/tmp/gh-aw/cache-memory/step-name-alignment/patterns.json
Priority
This issue is High Priority — 14 files affected, and the ambiguous naming makes it harder to understand workflow structure at a glance.
AI generated by Step Name Alignment for daily maintenance
Generated by Step Name Alignment · ◷
Step Name Alignment Issues
Found in: multiple
.lock.ymlfilesSummary
Several workflows use
Cache <X>as a step name for cache-save operations. The established project pattern (as used forSave qmd models to cache,Save qmd index to cache,Save cache-memory to cache (default)) isSave <X> to cache. TheCache <X>naming is ambiguous — it doesn't clarify whether data is being saved or restored — and is inconsistent with the rest of the codebase.Issues Identified
[High Priority] Inconsistent pattern: "Cache (X)" → "Save (X) to cache"
Affected step names and files:
Cache node-llama-cpp binariesSave node-llama-cpp binaries to cachedaily-doc-healer,daily-doc-updater,dev,developer-docs-consolidator,dictation-prompt,glossary-maintainer,smoke-codex,technical-doc-writer,unbloat-docs,weekly-blog-post-writer(10 files)Cache layout spec dataSave layout spec data to cachelayout-spec-maintainer(1 file)Cache prompt clustering dataSave prompt clustering data to cacheprompt-clustering-analysis(1 file)Cache downloaded dataSave downloaded datadaily-news(1 file — uses repo-memory, notactions/cache)Cache cache-memory file share dataRestore cache-memory file share dataai-moderator(1 file — this step restores data, not saves it)Issue:
All of these steps either use
actions/cache@...to save data to the GitHub Actions cache, or perform a custom save to repo-memory. The established naming pattern (Save <X> to cache) makes the save direction explicit. The ambiguousCache <X>prefix is used correctly for save operations inqmd-related steps elsewhere, and these should match.Note:
Cache cache-memory file share datainai-moderator.lock.ymlis particularly confusing because it is actually a restore operation, not a save.Established pattern references:
Save qmd models to cache✅Save qmd index to cache✅Save cache-memory to cache (default)✅Agentic Task Description
To fix these step names:
.mdsource files (not the.lock.ymlfiles directly):Cache node-llama-cpp binaries→Save node-llama-cpp binaries to cache(in all 10 workflows)Cache layout spec data→Save layout spec data to cache(inlayout-spec-maintainer.md)Cache prompt clustering data→Save prompt clustering data to cache(inprompt-clustering-analysis.md)Cache downloaded data→Save downloaded data(indaily-news.md)Cache cache-memory file share data→Restore cache-memory file share data(inai-moderator.md)gh aw compile <workflow>.mdfor each changed source file.lock.ymlfiles reflect the new namesRelated Files
.github/workflows/<name>.lock.yml/tmp/gh-aw/cache-memory/step-name-alignment/patterns.jsonPriority
This issue is High Priority — 14 files affected, and the ambiguous naming makes it harder to understand workflow structure at a glance.