Audit MCP tool contracts and compatibility; includes owned-service HTTP check demos.
Scored across 4 tools
The set forms a clean 2x2 of {demo, live} x {run check, read history}, and each description explicitly excludes the likely misinterpretation (e.g. 'NOT an external service probe', 'Not third-party monitoring'). An agent can select unambiguously between synthetic vs real checks and between executing a check vs reading stored results.
Names follow a predictable scheme: check_<mode> for actions and <mode>_history for reads, with the demo/live qualifier used consistently across both pairs. The verb_noun and noun_noun forms differ slightly, but the shared qualifier makes the pattern immediately readable.
Four tools is a tight, well-scoped surface for an experimental health-check server covering both synthetic and real paths plus their histories. It is on the lean side — nothing extraneous, but a small set for anything beyond this narrow purpose.
Each mode has a complete run-then-read loop (check_demo/demo_history, check_live/live_history), so no dead ends exist for the stated purpose. Minor gaps: no per-check detail lookup, no aggregate summary across the two modes, and no way to clear or reset demo data.
Audit supplied MCP tools/list descriptors and compare with optional previous_tools to flag likely breaking changes. No network requests or storage of your schemas. Returns structured findings and a SHA-256 fingerprint.
Inputs: tools (required), previous_tools.
Run a controlled synthetic health check and store its result. NOT an external service probe.
Inputs: scenario (required).
Perform a real HTTP check ONLY against Guardian's own fixed test target. Globally rate-limited to one check per minute. Not third-party monitoring.
Inputs: no named parameters.
Read measured synthetic check history over the last 24 hours. NOT independent uptime monitoring.
Inputs: scenario.
Discover project-owned complementary MCP tools useful for a specific agent task. No automatic call, login, payment or referral tracking.
Inputs: need.
Read real HTTP observations of Guardian's own test target from the last 24 hours, including sample size, availability, and latency.
Inputs: no named parameters.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Audits any remote agent or MCP URL for JSON-RPC compliance, latency SLA, OpenAPI schema, and HTTP.
Audit, pin and diff agent tool lists; preflight and watch remote MCP servers. Free tier; x402.
MCP deployment checks: required tools, schema drift and verifiable receipts. Free preflight.
Audit any domain or agent for A2A, MCP, x402, MPP, ERC-8004, and Poppy readiness.
Add one secure layer between your agents and this server.
