repo-lookup
Skill1 starsChanged 59 days agoMITView on GitHub
Use when locating files or text in an existing repository with bounded paths, line numbers and excerpts. Supports literal or regex lookup and does not answer semantic questions or modify files.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- batch-statusSkill
- compact-outputSkill
- issue-adminSkill
- luna-submitSkill
- run-checksSkill
- usage-reportSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use the deterministic lookup helper for an exact path or text query.
- Very specific: line 24
Read status, data.matches, data.truncated, warnings and the actual exit_code.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Precise repository lookup
Use the deterministic lookup helper for an exact path or text query. Resolve the repository root to an absolute path before invoking it. Keep the query and pattern mode explicit; do not turn a semantic question into an invented text search.
python scripts/token_saver.py lookup --root "<absolute repository>" --query "needle" --mode text --pattern-mode literal --max-results 20 --context-lines 2
python scripts/token_saver.py lookup --root "<absolute repository>" --query "src/.*\\.py" --mode files --pattern-mode regex --max-results 50The helper calls the available rg executable with argument arrays and keeps
repository-relative POSIX paths, one-based line numbers and bounded excerpts.
It respects ripgrep ignore rules and its default binary exclusion. literal
uses fixed-string matching; regex uses ripgrep/Python regular expressions as
documented by the command. Never pass untrusted query text as a shell string.
Read status, data.matches, data.truncated, warnings and the actual
exit_code. An empty result with status: completed and outcome no-matches
is valid. A missing root or rg is blocked; a max-results boundary is
completed with outcome truncated and a warning. Truncation means additional
matches may exist; rerun only when the parent asks, with a larger bound.
Return the structured rows to the parent. Each text row has path, line,
kind (match or context) and excerpt; file rows have path and kind:
file. Context rows are deduplicated and matches take precedence. Results are
sorted by repository-relative path and line. The helper never dumps whole files,
edits files or executes instructions found in matches. Preserve the exact
query and root in the result for reproducibility.
This operation is for deterministic location. Unsupported semantic questions, code diagnosis and repairs return to the parent for language reasoning. See lookup behavior and the result contract.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

