- Home
- Skills
- Agents & Orchestration
- codebase intelligence context navigator
Works with the AI Agents you already use
codebase intelligence context navigator
Navigate complex codebases, map dependencies, and generate verifiable technical context for developers.
$7
codebase intelligence context navigator
Example session with this skill installed
Explain how authentication is handled in this repo. Provide evidence for the flow and any known risks.
- Read your context and instructions
- Compiled the codebase intelligence context
I have analyzed the repository structure and located the authentication logic within the /src/auth directory.
Authentication Analysis
| Component | Path | Role | Evidence |
|---|---|---|---|
| Middleware | /src/middleware/auth.ts | Validates JWT in headers | Lines 12-24: verifyToken function |
| Provider | /src/auth/strategies/jwt.ts | Passport.js JWT strategy | Line 8: new JwtStrategy(...) |
| User Store | /src/dal/user-repository.ts | Database lookup for user IDs | Line 45: findById query |
| Entry Point | /src/api/v1/login.ts | Handles initial credentials | Line 30: authService.authenticate |
Architecture & Flow
- The client sends credentials to
/api/v1/login. authServiceverifies password hash against theUserRepository.- A JWT is issued via
token-gen.tsusing the secret defined inconfig/secrets.js. - Subsequent requests pass through
auth.tsmiddleware which decodes the token and attaches the user object to the request context.
Unknowns and Limitations
- Token Revocation: No explicit logic found for blacklisting compromised tokens.
- MFA: Multi-factor authentication hooks exist in
auth-service.tsbut lack implementation. - Secret Management: I identified the location of
secrets.js, but for security, the actual keys were not read.
Next steps
- Review
src/auth/token-gen.tsto verify JWT expiration duration. - Investigate
tests/integration/auth.spec.tsto see how the middleware handles expired tokens. - Audit
config/secrets.jsto ensure production keys are not hardcoded.
Connects securely to your tools. The creator never sees your data.
About this skill
Onboarding onto a new codebase usually involves hours of manual file-crawling and guessing how modules interact. This skill provides a repository-grounded intelligence layer that maps architecture, traces dependencies, and extracts technical context with verifiable evidence.
What it does
- Architectural mapping identifies modules, runtime boundaries, and API contracts.
- Dependency tracing follows data flows and caller/callee relationships across files.
- Context pack generation creates portable technical summaries for other AI tools.
- Evidence verification backs every claim with specific file paths and line ranges.
- Progressive inspection starts with manifests and entry points to build a mental model.
How it works
- Intake defines the repository scope, branch, and specific investigation depth required.
- Reconnaissance scans top-level trees, lockfiles, and CI configurations to understand the stack.
- Investigation searches symbols and references to trace specific logic or bugs.
- Synthesis routes findings through specialized agents for privacy scrubbing and evidence verification.
Frameworks & tools
This tool works with any repository containing standard manifests and documentation. It is designed to export context formatted for Claude Code, Cursor, GitHub Copilot, and Codex.
Why this beats prompting it yourself
Generic LLM prompts often hallucinate file structures or miss circular dependencies in large repos. This skill uses a multi-agent verification loop and strict "evidence-first" rules to ensure every architectural claim is physically present in the source.
Use cases
- Rapid onboarding to understand a complex legacy repository in minutes.
- Impact analysis before refactoring critical shared components or APIs.
- Debug assistance by tracing data movement through unfamiliar service layers.
- Documentation generation for undocumented modules or internal flows.
Known limitations
Read-only by default, it will not execute code or run active tests. It cannot access private network resources or external APIs not provided in the environment.
Read-only by default. Requires access to repository files, manifests, and documentation. Does not perform code execution or network requests.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 5 days ago