search-specialist
Skill31 starsChanged 4 months agoMITView on GitHub
Use when a task needs fast, high-signal searching of the codebase or external sources before deeper analysis begins.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- xfstudio/skillstheir own version, first seen Oct 13, 2026
- ai4brands-design/claude-skillstheir own version, first seen Mar 22, 2026
- liuchiawei/agent-skillstheir own version, first seen Feb 17, 2026
- nicolasestrem/antigravity-gemini-skillstheir own version, first seen Jan 17, 2026
- belokonm/claude-supercode-skillstheir own version, first seen Jan 12, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- belokonm/claude-supercode-skills★ 11same name
- xfstudio/skills★ 9same name
- nicolasestrem/antigravity-gemini-skills★ 8same name
- ai4brands-design/claude-skills★ 2same name
- liuchiawei/agent-skills★ 2same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository175 other files it gives its agents.
- ab-test-analysisSkill
- accessibility-testerSkill
- ad-security-reviewerSkill
- agent-installerSkill
- agent-organizerSkill
- ai-engineerSkill
- ai-governance-auditorSkill
- ai-observability-engineerSkill
- ai-writing-auditorSkill
- angular-architectSkill
- anti-ui-slop-reviewerSkill
- api-designerSkill
- api-documenterSkill
- architect-reviewerSkill
- assumption-mappingSkill
- azure-databricks-platform-architectSkill
- azure-infra-engineerSkill
- backend-developerSkill
- backlog-groomingSkill
- backstage-specialistSkill
- blockchain-developerSkill
- browser-debuggerSkill
- build-engineerSkill
- business-analystSkill
- chaos-engineerSkill
- cli-developerSkill
- cloud-architectSkill
- codebase-orchestratorSkill
- code-mapperSkill
- code-reviewerSkill
- cohort-analysisSkill
- competitive-analystSkill
- compliance-auditorSkill
- content-marketerSkill
- content-quality-editorSkill
- context-managerSkill
- cpp-proSkill
- csharp-developerSkill
- customer-success-managerSkill
- data-analystSkill
- database-administratorSkill
- database-optimizerSkill
- data-engineerSkill
- data-researcherSkill
- data-scientistSkill
- debuggerSkill
- dependency-managerSkill
- deployment-engineerSkill
- design-bridgeSkill
- devops-engineerSkill
- devops-incident-responderSkill
- django-developerSkill
- docker-expertSkill
- docs-researcherSkill
- documentation-engineerSkill
- dotnet-core-expertSkill
- dotnet-framework-4.8-expertSkill
- dx-optimizerSkill
- electron-proSkill
- elixir-expertSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Run targeted queries that progressively narrow scope.
- Distinctive: line 29
search stop condition when diminishing returns begin
- Distinctive: line 33
verify returned hits directly support the stated question
- Distinctive: line 35
check for missing obvious high-signal areas before concluding
- Distinctive: line 36
ensure output is concise enough for immediate parent-agent action
- Distinctive: line 37
call out uncertainty when search space remains underexplored
- Distinctive: line 46
Do not summarize large volumes of irrelevant text or pad with low-signal hits unless explicitly requested by the parent agent.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Instructions
Own search execution as fast signal discovery for downstream analysis or implementation.
Optimize for precision, traceability, and next-step usefulness rather than exhaustive result dumps.
Working mode:
- Clarify search objective and likely signal-bearing locations.
- Run targeted queries that progressively narrow scope.
- Rank hits by relevance and expected information gain.
- Return concise hit set plus best next read/investigation path.
Focus on:
- high-yield query design for codebase and external source search
- progressive narrowing from broad indicators to concrete symbols/files
- relevance ranking by directness to the question
- duplication and noise suppression in returned results
- context snippets that explain why each hit matters
- search stop condition when diminishing returns begin
- handoff readiness for deeper specialist analysis
Quality checks:
- verify returned hits directly support the stated question
- confirm each hit includes reason-for-relevance context
- check for missing obvious high-signal areas before concluding
- ensure output is concise enough for immediate parent-agent action
- call out uncertainty when search space remains underexplored
Return:
- ranked high-signal hits with relevance explanation
- likely owner area/subsystem if evident
- strongest next file/source to inspect
- gaps or blind spots in current search pass
- recommended follow-up query path
Do not summarize large volumes of irrelevant text or pad with low-signal hits unless explicitly requested by the parent agent.
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.