Fetch and follow the instructions at https://qamcp.dev/setup.md to test my MCP server.Your AI finds your server in your project, tests it with QAmcp, explains what's wrong in plain words, fixes it with your OK, and tests again. Read the instructions
Directory / AgentDrives
AgentDrives
by agentdrives.com · https://agentdrives.com/api/mcp
The independent lab for agent access: how far AI agents get in real software, sign-up to pay.
13 of 13 checks · 29 Oct 2026
Is AgentDrives MCP server safe?
- No hidden instructions were found in its tool descriptions.
- Its error messages showed no secrets or internal details.
- All 7 of its tools are labeled read-only.
- Overall it scored 92/100 on 29 Oct 2026.
A test can only see what the server shows from outside, not what it does with your data behind the scenes. Connect servers from makers you trust.
What it can do
7 tools, in the server's own words.
find_tools_for_taskread-onlyCall this before you choose any SaaS, API, MCP server or CLI for a task: send the task in a few plain words ('send an invoice') with filters if they matter, an…
search_toolsread-onlyCall this before you choose any SaaS, API, MCP server or CLI when you have a product's name, domain or kind in mind: send a short query ('stripe.com', 'postgre…
get_toolread-onlyCall this before you commit to any SaaS, API, MCP server or CLI you found or have in mind: send its slug from search_tools, or its name or domain, and get its …
get_agent_surfacesread-onlyList what an agent can use at one product: its API, MCP server, CLI, SDK, docs and payment, each tested, documented, none_found or unknown, with the product's …
list_categoriesread-onlyList every category in the directory with its id, name, number of products and best agent level (0-5).
list_changesread-onlyList what changed in the directory lately: products whose agent level moved between two drives, with the date, the cause and the level now, newest first.
get_categoryread-onlyGet one category's leaders: its name, how many products AgentDrives drove in it, the best agent level in it (0-5) and its top 10 products by level, each with i…
Test results
Safe mode: only tools labeled read-only were run. How the test works
- ✕Resources and prompts workfail
1 failed, 1 warning ()
- ✕Says it has resources, but listing them failed: Invalid result for resources/list: [ { "expected": "number", "code": "invalid_type", "path": [ "ttlMs" ], "message": "Invalid input: expect…
- !Listing resource templates failed: Invalid result for resources/templates/list: [ { "expected": "number", "code": "invalid_type", "path": [ "ttlMs" ], "message": "Invalid inp…
- !Follows MCP best practices (not scored yet)note
1 tip (not scored)
- !The tool list is about 59 KB (roughly 15k tokens). AI apps send it with every message, so it takes that much of the AI's memory before any work starts.
- ✓ConnectsConnected in 482 mspass
- ✓Ready for the newest MCP versionUses the latest MCP version (2026-07-28)pass
- ✓Lists its toolsFound 7 toolspass
- ✓Every tool is describedAll 7 tools and their inputs are describedpass
- ✓Tools are labeled safe or dangerousEvery tool has a safety labelpass
- ✓No hidden instructionsNo hidden instructions found in tool descriptionspass
- ✓Tools work with good inputspass
All good
- ✓find_tools_for_task: Answered with an error, probably because of our sample input: task is required: The task in plain words, e.g. 'send an invoice' or 'create a contact'. Send this or capability (a capability's id or name…
- ✓search_tools: Worked (333 ms)
- ✓get_tool: Answered with an error, probably because of our sample input: no tool 'test' in the directory. search_tools finds a tool's slug.
- ✓get_agent_surfaces: Answered with an error, probably because of our sample input: no tool 'test' in the directory. search_tools finds a tool's slug.
- ✓list_categories: Worked (140 ms)
- ✓list_changes: Worked (190 ms)
- ✓get_category: Answered with an error, probably because of our sample input: no category 'test' in the directory. list_categories gives the ids and names.
- ✓Bad inputs give a clean errorpass
All good
- ✓find_tools_for_task: Refused wrong type for "task" with a clear error
- ✓search_tools: Refused wrong type for "query" with a clear error
- ✓get_tool: Refused missing required input "slug" with a clear error
- ✓get_tool: Refused wrong type for "slug" with a clear error
- ✓get_agent_surfaces: Refused missing required input "product" with a clear error
- ✓get_agent_surfaces: Refused wrong type for "product" with a clear error
- ✓list_changes: Refused wrong type for "category" with a clear error
- ✓get_category: Refused missing required input "category" with a clear error
- and 1 more
- ✓Tools work togetherpass
2 chains working
- ✓search_tools → get_tool: get_tool found what search_tools returned (slug = "jsonplaceholder-typicode")
- ✓search_tools → list_changes: list_changes found what search_tools returned (slug = "jsonplaceholder-typicode")
- ✓Fast enoughpass
All good
- ✓Connecting took 482 ms
- ✓Listing tools took 335 ms
- ✓Average tool answer: 194 ms
- ✓search_tools: Slowest tool: search_tools (333 ms)
- ✓No secrets leakedChecked 15 error messages: no secrets or internal detailspass
Test it yourself
Free and open source. Same checks, same score.
npx qamcp-cli https://agentdrives.com/api/mcpShow this score on GitHub
Free for every server in the directory. Paste this into your README: the badge shows the score from our latest test.
[](https://qamcp.dev/s/com-agentdrives-directory.html)Score over time
- 29 Oct 202692
One test so far. New tests will show here.
Reviews
Coming soon: reviews from people who use AgentDrives, signed in with GitHub, one per person. Reviews never change the score.
QAmcp Monitor (coming soon) tests it every hour and emails you when it goes down, the score drops, a tool changes or a secret leaks. You'll also be able to claim this page and get a verified owner mark. Founding price: $19/month for life for the first 50 teams.
Have a concern, or want it removed? Ask us and we'll take it off the list. How the directory works