Works with the AI Agents you already use
MCP Contract Guard Review
Catch changes to MCP tool descriptions, schemas and annotations before granting access. Compare exports locally, explain what needs review and produce JSON or SARIF evidence for CI.
$10
MCP Contract Guard Review
Example session with this skill installed
Compare the bundled candidate catalog against the reviewed baseline. Should this MCP update proceed without another review?
- Read your context and instructions
- Compiled the mcp contract guard
Review required. The local CLI exited with code 2. In this synthetic catalog, one tool was added, one removed, four changed and one stayed unchanged. The report contains six surface signals and one advisory text finding.
record.archive: The schema no longer requires confirmation and the destructive-action hint changed. Ask where confirmation is enforced.report.export: The schema added a destination and broadened accepted inputs. Verify which destinations the implementation actually accepts.record.bulk_updatewas added andlegacy.lookupwas removed. Review the new tool’s authorization and check callers of the removed tool.catalog.searchanddocument.preview: Review the changed descriptions and the advisory text signal in the preview description.
The unchanged control returned exit 0. An incomplete export returned exit 1 with no report. The changed export produced JSON, SARIF and HTML evidence.
Connects securely to your tools. The creator never sees your data.
About this skill
Know what changed before you approve an MCP server update.
A tool can keep its name while its description, inputs or declared behavior change. MCP Contract Guard Review gives your agent a reproducible way to compare a candidate tool catalog against a reviewed reference and tell you what needs a second look.
- Compare complete
tools/listexports with the local Contract Guard CLI. - See additions, removals, schema and annotation changes, and advisory text findings.
- Get a review handoff with evidence paths and questions for the tool owner.
- Use the same decision in CI and export JSON, SARIF or HTML.
The skill includes the review workflow, setup and CI guidance. The comparator is available separately from the project’s public source. Your team decides when to approve a baseline.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.
Compares complete exported tool definitions against a reviewed baseline. Requires Node.js 22+ and the local CLI. It cannot verify live behavior, authorization, or the completeness of the export.
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 buying a skill.
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
- Software
Node.jsMCP Contract Guard CLIContract Guard MCP
Git
- Optional
- Git for first-time setup
Permissions
Reads only the selected baseline and candidate JSON files and writes a new report when requested. Comparisons run locally. First-time setup fetches pinned source and dependencies from GitHub and npm.