- Home
- Skills
- Code Quality & Review
- Ecosystem Integration Review for Repositories
Ecosystem Integration Review for Repositories
Get evidence-backed findings on framework and library integrations, with exact locations and testable fixes.
Free
Works with the AI Agents you already use
Ecosystem Integration Review for Repositories
Example session with this skill installed
Review this Python repo's FastAPI/SQLAlchemy boundary. app/db.py:18 creates a session per request; lines 19-20 yield it without cleanup. No lockfile or tests are supplied.
- Read your context and instructions
- Compiled the ecosystem integration review
Reviewed the supplied session-handling evidence. One lifecycle concern needs verification; no files were changed.
Discovery and scope
Python, FastAPI, and SQLAlchemy are identified from your description. Versions are unverified because no lockfile was supplied.
| Boundary | Priority | Evidence available |
|---|---|---|
| FastAPI dependency to SQLAlchemy session | High | Reported creation and yield at app/db.py:18-20 |
Finding
| Location | Observed pattern | Impact | Confidence | Recommendation |
|---|---|---|---|---|
app/db.py:18-20 | A per-request session is yielded without cleanup in the described code. | If nothing else closes it, database resources may remain held beyond the request. | Provisional. Surrounding code and lifecycle ownership are unavailable. | Find the session owner. If cleanup is absent, add exception-safe closure and test it. |
Boundary and evidence limits
The request-to-session boundary was assessed from your description, not direct file inspection. External cleanup remains a possible explanation. No tests were supplied or executed, and version-specific behavior was not verified against official sources.
Next steps
- Inspect the complete dependency and registration. Acceptance: identify who owns session cleanup.
- Add success and exception-path tests. Acceptance: verify closure after both request outcomes.
- If cleanup is missing, propose a focused patch. Acceptance: the lifecycle tests pass without unrelated changes.
Connects securely to your tools. The creator never sees your data.
About this skill
A dependency list cannot tell you whether your frameworks and libraries work together correctly. Get an evidence-based review of how your repository actually uses them, with specific findings and small, testable recommendations.
What it does
- Finds actual usage: Examines source code, configuration, build files, and tests rather than just installed packages.
- Reviews integration boundaries: Follows how technologies handle requests, database access, background work, external clients, and runtime setup.
- Supports mixed-language repositories: Identifies materially used languages without assuming the top-level project tells the whole story.
- Explains actionable findings: Reports exact locations, likely impact, confidence, uncertainty, and a focused recommendation.
- Checks version-sensitive claims: Consults official documentation when available and flags anything it cannot verify.
How it works
- Provide repository access and ask for an integration review.
- Optionally name languages or technologies to prioritize and choose whether to allow web research.
- Receive a scoped report with findings, evidence limits, and follow-up requests you can approve separately.
Works with
Repositories in any language, including projects that combine languages. Reviews libraries, frameworks, runtimes, build systems, databases, messaging tools, testing tools, and SDK integrations where repository evidence supports the assessment.
Why use this skill
Spend less time sorting dependency lists into useful review work. Recommendations are checked against available code, configuration, tests, and official sources to reduce speculative advice.
Use cases
- Review database session handling in a web application.
- Check integration assumptions before a framework migration.
- Examine messaging boundaries across a mixed-language service.
Known limitations
Read-only by default. Not a broad architecture, style, security, or performance audit. Missing repository or source evidence limits conclusions.
Read-only unless changes are authorized. Excludes broad architecture, style, security, and performance audits. Missing repository evidence or official sources limits confidence.
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
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 month ago
- Free with an Agensi account
Permissions
Reads local manifests, source, configuration, tests, and lock files; may run read-only inspection commands. Browser/network access is used only for official version-sensitive documentation when available and allowed. The default review does not edit repository files.