MCP inspection / Code and repository research
Grep MCP
See what passed, what is missing, and what to test next.
Anonymous endpoint inspection
·Runner 1.1.04/4
Metadata checks passed
Inspection overview
Metadata checks passed
The endpoint connected and its advertised tool definitions passed the recorded checks.
1
Tools discovered
Complete catalog
-
Read-only tool call
Not tested
Check breakdown
Select a category to inspect its findings.
Connection2/2Protocol handshake and initialization
Passed
Protocol handshake and initialization
Passed
- Protocol negotiationPassed
Negotiated protocol 2025-06-18; this does not establish host compatibility.
View request evidence - InitializationPassed
The initialization notification received HTTP 202 with an empty body.
View request evidence
Tool discovery1/1Can the endpoint list its tools?
Passed
Can the endpoint list its tools?
Passed
- Catalog discoveryPassed
Observed the complete advertised catalog: 1 tools across 1 pages.
View request evidence
Metadata shape1/1Tool definitions accepted by the SDK
Passed
Tool definitions accepted by the SDK
Passed
- Metadata structurePassed
SDK metadata shape checked for 1 observed tools. This is not complete JSON Schema validation or behavior verification.
View request evidence
Tool metadata coverage
Field presence across 1 observed tool. Presence does not measure quality; output schemas and behavior hints are optional.
1 of 1 recorded
Tool-by-tool details
Character count only; wording has not been assessed.
- searchGitHub1537 characters
1 of 1 recorded
Tool-by-tool details
Root shape only; full schema validity has not been assessed.
- searchGitHubtrue
0 of 1 recorded
Tool-by-tool details
An optional output contract; matching responses requires execution testing.
- searchGitHubfalse
0 of 1 recorded · 1 unknown
Tool-by-tool details
A recorded true or false declaration counts as present. Behavior is unverified.
- searchGitHubNot recorded
0 of 1 recorded · 1 unknown
Tool-by-tool details
A recorded true or false declaration counts as present. Behavior is unverified.
- searchGitHubNot recorded
Request timings
Single observations from this run, not a performance benchmark.
Test coverage
- OAuth
- Not tested
- ChatGPT
- Not tested
- Claude
- Not tested
- Overall quality grade
- Not graded
A passed metadata check does not establish login, client compatibility or task reliability.
What to do next
Consider output schemas
For tools with structured results, an output schema can make the response contract explicit. This field is optional.
Affected tools: searchGitHub
Review behavior hints
Check the source definitions and declare read-only or destructive behavior accurately. These optional hints are not verified behavior.
Affected tools: searchGitHub
Test a real client workflow
Choose a task and expected result, then verify it in the client your users depend on.
Run details and raw evidence
Requests, inputs, responses, catalog, provenance and test limits
Run details and raw evidence
Requests, inputs, responses, catalog, provenance and test limits
Recorded checks
These checks cover the metadata inspection. They are separate from the tool-call result above, when present.
protocol-negotiation
Negotiated protocol 2025-06-18; this does not establish host compatibility.
initialized-notification
The initialization notification received HTTP 202 with an empty body.
tool-metadata-shape
SDK metadata shape checked for 1 observed tools. This is not complete JSON Schema validation or behavior verification.
oauth-login
OAuth discovery, login, token refresh and authenticated access are outside this anonymous inspection.
Metadata request evidence
request-1 / initialize
2026-10-15T07:14:20.980Z
HTTP 200 · 643 ms · 209 B
response-received
request-2 / notifications/initialized
2026-10-15T07:14:21.623Z
HTTP 202 · 286 ms · 0 B
notification-accepted
request-3 / tools/list
2026-10-15T07:14:21.909Z
HTTP 200 · 236 ms · 3043 B
response-received
Observed tool catalog
Names were advertised through tools/list. Any tool execution is recorded separately in the smoke test section.
Complete: 1 tool
- searchGitHub
Endpoint and provenance
- Endpoint
- https://mcp.grep.app
- Official source
- https://vercel.com/blog/grep-a-million-github-repositories-via-mcp
- Profile
- Anonymous endpoint profile
- Requested protocol
- 2025-11-25
- Reported protocol
- 2025-06-18
- Protocol
- 2025-06-18
- Observed at
- 2026-10-15T07:14:22.145Z
- Run ID
- 66b4bd66-b993-496b-8738-5f67e89fddc4
- Catalog SHA-256
- cd321be5eb52de5d2a312cb0f26c68434a56c03ebb6519627ac07a4c5b79c52d
- Fingerprint scope
- complete-catalog
SHA-256 of all observed tool objects, recursively sorted object keys and sorted canonical tool strings; schema array order is preserved. Raw descriptions, schemas and headers are not persisted.
Inspection bounds
- requestTimeoutMs
- 5,000
- runTimeoutMs
- 30,000
- responseBytes
- 1,048,576
- requests
- 7
- pages
- 5
- tools
- 200
What this report does not establish
These observations do not establish general task success, security, OAuth behavior or ChatGPT and Claude compatibility. Read the limits of each recorded test.
- Metadata-only anonymous inspection. No tools/call, OAuth flow, API model calls, host UI or publication review was executed. No global quality or safety score is produced.
- Annotations are server declarations, not verified behavior. Metadata success does not establish task success, security, usefulness or ChatGPT/Claude compatibility.
- Only active exact HTTPS endpoints in the reviewed source registry are accepted. Redirects, supplied credentials, arbitrary endpoints and OAuth URL fetching are disabled. DNS/TLS use the local runtime: this is not a hardened multi-tenant SSRF sandbox and must not be exposed as a public scanner.
- Session identifiers are used only in memory. Request/response headers, raw error messages, raw descriptions and schemas are excluded from artifacts. Tool names and aggregate metadata remain untrusted server input.
- Only the recorded protocol versions and bounded catalog prefix are supported. Reaching a bound or lacking access means unknown, not a failed server or a passing full audit.
Compare actual extraction results
Our web-to-Markdown benchmark tests shared pages and shows the content checks behind every result.
View the benchmark