agentleFS
tkollehtkolleh/skills/sourcegraph-search/SKILL.md

sourcegraph-search

Skill1 starsChanged 3 months agoMITView on GitHub

Search code, commits, and diffs across millions of open source and private repositories using the Sourcegraph CLI (src).

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 14
  2. Very specific:

    Use patternType:structural to match code structures rather than exact strings.

    line 45
  3. Very specific:

    Syntax: Use :[name] as a placeholder (hole) to match any code between delimiters.

    line 47
  4. Distinctive:

    Use these specific patterns for high-value tasks:

    line 57
  5. Very specific:

    Check package.json files for a specific vulnerable library version.

    line 76

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

What I do

I help you perform large-scale code analysis and discovery by searching Sourcegraph. I use the src CLI to execute precise queries that can match literal strings, regular expressions, or structural code patterns. I can filter results by repository, file type, language, commit history, and time ranges. I am particularly effective at finding usage examples, security vulnerabilities, deprecated patterns, and architectural inconsistencies across massive codebases.

When to use me

Use this skill when:

  • You need to search across a large codebase or multiple repositories for specific patterns, functions, or dependencies.
  • You need to find real-world usage examples of a function or API across many repositories.
  • You are auditing a codebase for security tokens, keys, or vulnerable dependencies.
  • You want to perform "structural search" to match code patterns regardless of whitespace or formatting (e.g., finding all try-catch blocks with empty catches).
  • You need to track recent changes or history using type:diff or type:commit (e.g., "what changed in the last week").
  • You want to find projects that use specific dependencies (e.g., "who relies on lodash version 4.17.19").
  • You need to identify repositories that are contributor-friendly (e.g., contain CONTRIBUTING.md).

Instructions

1. Basic Search

Execute searches using the src search command. Always use the -json flag to ensure the output is machine-readable and easy to parse. Enclose your query in single quotes to prevent shell expansion.

src search -json 'query'

2. Advanced Filtering

Refine your search results using these standard filters within your query string:

  • Repository: repo:^github\.com/owner/name$ (supports regex)
  • File: file:\.js$ or file:package\.json
  • Language: lang:TypeScript
  • Content: content:"string to find" (optional, content is searched by default)
  • Boolean Operators: AND, OR, NOT (e.g., error AND NOT file:test)

3. Structural Search

Use patternType:structural to match code structures rather than exact strings. This is powerful for finding code regardless of line breaks or formatting.

  • Syntax: Use :[name] as a placeholder (hole) to match any code between delimiters.
  • Example: Find console.log with any argument:
src search -json 'console.log(:[args]) patternType:structural'

4. Search Tricks & Best Practices

Use these specific patterns for high-value tasks:

Find Usage of a Method (Structural) Find where a method is called, even with different arguments or formatting.

src search -json 'myMethod(:[args]) patternType:structural'

Find Security Vulnerabilities (Regex) Scan for potential leaked secrets or keys.

src search -json '(key|secret|token)-[\w+]{32,} patternType:regexp'

Find Specific Dependency Versions Check package.json files for a specific vulnerable library version.

src search -json 'file:package.json lodash 4.17.19'

Find Recent Changes (Diffs/Commits) See what changed in a repository recently.

src search -json 'repo:^github\.com/owner/repo$ type:diff after:"1 week ago"'

Find Contributor-Friendly Projects Locate repositories with contribution guidelines.

src search -json 'contributing lang:Markdown'

5. Output Interpretation

The -json output will provide a list of matches. Key fields to observe:

  • repository: The repository name.
  • file: The path to the matching file.
  • lineMatches: Specific lines and line numbers where the match occurred.
  • commit: If searching history, the commit hash and author.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.