2
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill fusengine/agents --skill exploration- SKILL.md7.5 KB
Overview
This skill provides a repeatable codebase exploration protocol for rapid discovery, architecture analysis, pattern detection, and dependency mapping. It is tuned for TypeScript and polyglot repositories and produces evidence-based summaries that guide next steps for refactoring, onboarding, or security review. The skill emphasizes lightweight commands and clear indicators so you can get reliable insights within minutes.
How this skill works
It runs a phased inspection: reconnaissance to surface root files and config, structure mapping to build a directory map, entry-point detection to locate executables and servers, dependency analysis to extract runtime/dev libs, and pattern detection to infer architectural style. The outputs combine simple shell probes (ls, grep, tree, find) with heuristic indicators (directory names, key files) to produce a compact architecture assessment and recommended actions.
When to use it
- Onboarding to a new repository to quickly locate entry points and tech stack
- Before planning a refactor to verify boundaries and where business logic lives
- During audits to detect large files, dependency risks, and missing tests
- When deciding if a project is monolith, modular monolith, or microservices
- To validate framework and ORM usage across mixed-language repos
Best practices
- Start with read-only commands and avoid changing files during discovery
- Prioritize evidence: quote matching files and grep hits when asserting patterns
- Limit tree depth and exclude noise folders (node_modules, dist, .git)
- Combine dependency file inspection (package.json, go.mod, pyproject) with grep for framework keywords
- Record entry points and top-level scripts before diving into implementation details
Example use cases
- Produce a one-page architecture summary for a TypeScript web app (tech, entry points, pattern)
- Map cross-repo dependencies in a monorepo to plan a modularization effort
- Detect violations like business logic inside UI components or oversized files
- Quickly confirm whether a service uses Next.js app router, Express, or Fastify
- Generate a prioritized checklist for security review based on dependencies and large files
FAQ
Detection is heuristic: it relies on directory and file indicators and grep hits. Confirm any architectural claim by inspecting the referenced files.
Can this run on non-TypeScript projects?
Yes. The protocol includes commands and indicators for Python, Go, Rust, and PHP as well as TypeScript.