بنقرة واحدة
compare
Compare duplicate research directories to determine which is superseded.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Compare duplicate research directories to determine which is superseded.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | compare |
| description | Compare duplicate research directories to determine which is superseded. |
| disable-model-invocation | true |
| argument-hint | [dir1] [dir2] |
| allowed-tools | ["Read","Glob","Grep"] |
Compare multiple Ralph loop directories that cover the same topic to determine which is superseded and whether anything needs porting.
/ralph:compare <dir1> <dir2> - Compare two directories/ralph:compare <dir1> <dir2> <dir3> - Compare three directories/ralph:compare - Will prompt for directories to compareGet directories to compare:
$ARGUMENTS provided, parse as space-separated directory pathsdocs/staged-learnings/tmp-feature docs/staged-learnings/feature)"Resolve content from local files or research branches:
spec.md + progress.md with no completed tasks): check for a research/<basename> branch via git branch -a --list *research/<basename>*git show <branch>:<path> and list files via git ls-tree --name-only <branch> <path>/research/<topic> branches — the local directory may only have initialization boilerplate while the branch has the full outputRead key files from each project (from whichever source resolved in step 2):
spec.md, progress.md, info.md, assumptions-and-questions.mdBuild comparison table: Using the checklist in comparison-checklist.md, compare:
| Aspect | Dir A | Dir B |
|--------|-------|-------|
| Total files | X | Y |
| Iterations tracked | N | M |
| Research lines (info.md) | ~200 | ~500 |
| Implementation phases | 6 | 10 |
| Has assumptions doc | No | Yes |
| Has iteration logs | No | Yes |
| Spec version | v1 | v2 |
Identify superseded directory: Check for signs of supersession:
Check for unique content to port:
Present recommendation:
## Comparison Summary
| Aspect | <dir1> | <dir2> |
|--------|--------|--------|
| ... | ... | ... |
## Recommendation
**Superseded**: <directory> is superseded by <directory>
**Porting needed**: <Yes/No>
<If yes, list specific sections>
**Action**: `rm -rf <superseded-directory>`
If consolidation needed instead (directories have complementary content):
## Recommendation
**Neither is fully superseded** - directories have complementary content.
**Consolidation approach**:
- Keep <directory> as target
- Port from <directory>:
- <file or section 1>
- <file or section 2>
- Delete <directory> after porting
User: /ralph:compare docs/staged-learnings/tmp-monte-carlo docs/staged-learnings/monte-carlo-simulation
Claude: Comparing directories...
## Comparison Summary
| Aspect | tmp-monte-carlo | monte-carlo-simulation |
|--------|-----------------|------------------------|
| Total files | 3 | 8 |
| Iterations tracked | 2 | 7 |
| Research lines (info.md) | ~150 | ~480 |
| Implementation phases | 4 | 8 |
| Has assumptions doc | No | Yes |
| Has iteration logs | No | Yes |
| Spec version | v1 | v2 |
## Recommendation
**Superseded**: tmp-monte-carlo is superseded by monte-carlo-simulation
**Porting needed**: No - monte-carlo-simulation covers all topics with more depth
**Action**: `rm -rf docs/staged-learnings/tmp-monte-carlo`
Capture session learnings and save to skills, guidelines, or reference docs under ~/.claude/.
Orchestrate parallel claude -p sessions — bootstrap, launch, monitor, and converge. Works with any skill that produces manifest.json, item directories, and a runner script.
Create a request (pull request or merge request) or update an existing one following project conventions.
Resolve merge or rebase conflicts between branches.
Assess open PRs with unaddressed review comments and generate a parallel addressing script — produces manifest.json and let-it-rip.sh for address-request-comments execution.
Assess open work items and generate a parallel execution script — produces manifest.json and let-it-rip.sh for implement/clarify execution.