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 / Agentalog — Agent Auto Log
Agentalog — Agent Auto Log
by agentalog.com · https://agentalog.com/mcp
Watch MCP tools, package publishers and domains. Free catalog; paid continuing checks.
13 of 13 checks · 10 Nov 2026
At a glance
- Runs where
- On their server. What you send to it goes to them.
- What you need
- Nothing: it answered without a login
- Tools
- 50 tools
- Its own safety labels
- 38 read-only, 1 labeled as deleting or destroying, 11 labeled as changing things
- Memory it takes
- About 13,000 tokens of your AI's memory, just for its tool descriptions
- Largest answer
- About 415.4 KB in our test
- Tools that look alike
- 2 pairs of tools with nearly the same description, so the AI may pick the wrong one
- Speed
- Average answer in our test: 240 ms
Tokens are an estimate: the characters in its tool names, descriptions and inputs, divided by 4. The exact number depends on your AI model. Safety labels are what the server says about itself. From our latest test, 10 Nov 2026.
Is Agentalog — Agent Auto Log MCP server safe?
- None of the hidden-instruction wordings we know were found in its 50 tool descriptions.
- Its error messages showed no secrets or internal details.
- 38 of its 50 tools are labeled read-only; the others may change or send things.
- Overall it scored 81/100 on 10 Nov 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 this test does not check
- Its source code and packages.
- What a tool quietly does behind the scenes, such as writing a file or calling another service.
- Instructions hidden inside live tool answers. We only look for those in tool descriptions.
- Hidden instructions worded in a way our list doesn't know yet.
- Tools that could change something. Those are read, never run.
What QAmcp can and can't see · How to tell if an MCP server is safe · Hidden instructions · Prompt injection
Known security advisories
Not looked up yet. QAmcp tests how a server behaves. It does not scan source code.
What it can do
50 tools, in the server's own words.
search_bundlesread-onlySearch the index of OKF bundles: free text over name, tagline, description and origin; filter by provenance (github|domain), repository, version, language, lic…
get_bundleread-onlyOne bundle's card: root index.md URL, version, concept count, provenance and repository signal.
index_statsread-onlySize of the index by provenance and when it last changed.
listing_dependenciesread-onlyRead a tool's declared package/repository/domain chain and cached public source evidence.
watch_dependency_changesUse an existing paid watch pass to follow package, publisher, repository and domain changes.
dependency_changesread-onlyRead your watched tool's before/after evidence, dates and sources.
feed_findread-onlyDoes this site publish a feed?
feed_postsread-onlyWhat has this feed published?
feed_postread-onlyRead one post whole: the body in Markdown (`content`) and the author's formatting as safe HTML (`content_html`), plus the link to the original.
feed_searchread-onlySearch the feed index by text and filters: format, audio/video, community, namespace and year of the newest item.
mcp_index_signalsread-onlyDoes this MCP server still answer, with which access, and is its repository alive?
mcp_index_searchread-onlyBrowse what independent MCP indexes measured, including servers this catalog does not list yet (casou=0).
check_toolread-onlyBefore installing a remote MCP server, check its address: signals from the address itself (consumer tunnel, no TLS), the catalog listings with that address or …
listing_securityread-onlyRead a remote MCP server or SKILL.md review before installing.
listing_toolsread-onlyThe tools and instructions a remote MCP server announced at its last check, with what they mention doing and any suspicious text.
list_watched_toolsread-onlyThe MCP servers and skills the account watches, with their review level, slots and plans.
watch_toolWatch a remote MCP or supported SKILL.md.
unwatch_toolStop watching a server; its slot becomes free.
list_plansread-onlyThe plans for sale, with price and term, and the account's active plan.
buy_planBuy a plan: vaga (1 more tool), equipe (50 more tools and up to 10 private projects), empresa (500 more tools and up to 10 private projects).
cancel_planCancel the active plan.
claude_code_allowlistread-onlyThe watch list as Claude Code managed settings (allowedMcpServers, deniedMcpServers) and claude mcp add commands.
watch_alertsread-onlyThe changes in the watched tools, oldest first, 100 per page.
a2a_agent_searchread-onlyFind A2A agents — agents that publish an AgentCard — by text, health, protocol version, registry, skill tag or publisher.
and 26 more
Test results
Safe mode: only tools labeled read-only were run. How the test works
- !Ready for the newest MCP versionnote
1 warning
- !Uses an older MCP version (2025-06-18). It works today, but it isn't on the July 2026 version (2026-07-28) yet.
- !Every tool is describednote
2 warnings
- !api_access and api_access_buy have nearly the same description (80% of their words are shared; the only difference: "monthly", "private", "purchase" vs "basic", "valid", "require"), so an AI may not know which to pick
- !feed_find and feed_search have nearly the same description (67% of their words are shared; the only difference: "does", "site", "answer" vs "audio", "video", "community"), so an AI may not know which to pick
- !Tools work with good inputsnote
1 warning
- –list_watched_tools: Needs you to sign in, so it was not tested (not counted against the server)
- –claude_code_allowlist: Needs you to sign in, so it was not tested (not counted against the server)
- –watch_alerts: Needs you to sign in, so it was not tested (not counted against the server)
- !api_index: Answered with about 415 KB of text. Many AI apps cut off tool answers this big, so the AI may never see the end of it.
- –api_usage: Needs you to sign in, so it was not tested (not counted against the server)
- –watch_dependency_changes: Not run: labeled as changing data
- –watch_tool: Not run: labeled as changing data
- and 33 more
- !Bad inputs give a clean errornote
10 warnings
- !search_bundles: Accepted wrong type for "q" without complaining
- !feed_find: Accepted missing required input "dominio" without complaining
- !feed_find: Accepted wrong type for "dominio" without complaining
- !feed_search: Accepted wrong type for "q" without complaining
- !mcp_index_signals: Accepted wrong type for "listing" without complaining
- !mcp_index_search: Accepted wrong type for "fonte" without complaining
- –watch_alerts: Needs you to sign in, so bad inputs were not tested
- !a2a_agent_search: Accepted wrong type for "q" without complaining
- and 32 more
- !Follows MCP best practices (not scored yet)note
3 tips (not scored)
- !50 tools don't say whether they reach outside services like the web (openWorldHint): search_bundles, get_bundle, index_stats, listing_dependencies, watch_dependency_changes, dependency_changes and 44 more
- !The tool list is about 57 KB (roughly 14k tokens). AI apps send it with every message, so it takes that much of the AI's memory before any work starts.
- !19 tools answer with structured data but don't declare its shape (outputSchema): search_bundles, index_stats, feed_find, feed_search, mcp_index_signals, mcp_index_search and 13 more. If those answers change shape, nothing can catch it; with
- ✓ConnectsConnected in 956 mspass
- ✓Lists its toolsFound 50 toolspass
- ✓Tools are labeled safe or dangerousEvery tool has a safety labelpass
- ✓No known hidden instructionsNone of the hidden-instruction wordings we know were found in tool descriptions. New wordings can get past this check.pass
- ✓Tools work togetherpass
5 chains working
- ✓search_bundles → listing_dependencies: listing_dependencies found what search_bundles returned (id = "okf_jacobpevans_ai_ai_llm_prompts_index_md_db1d93c4")
- ✓search_bundles → check_tool: check_tool found what search_bundles returned (url = "https://agentalog.com/api/bundles")
- ✓search_bundles → listing_security: listing_security found what search_bundles returned (id = "okf_jacobpevans_ai_ai_llm_prompts_index_md_db1d93c4")
- ✓search_bundles → get_listing: get_listing found what search_bundles returned (id = "okf_jacobpevans_ai_ai_llm_prompts_index_md_db1d93c4")
- ✓search_bundles → get_listing_readme: get_listing_readme found what search_bundles returned (id = "okf_jacobpevans_ai_ai_llm_prompts_index_md_db1d93c4")
- ✓Fast enoughpass
All good
- ✓Connecting took 956 ms
- ✓Listing tools took 340 ms
- ✓Average tool answer: 240 ms
- ✓api_index: Slowest tool: api_index (751 ms)
- ✓No secrets leakedChecked 75 error messages: no secrets or internal detailspass
- ✓Resources and prompts workOpened 6 of 6 promptspass
Test it yourself
Free and open source. Same checks, same score.
npx qamcp-cli https://agentalog.com/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-agentalog-meta-agent-tools.html)Score over time
- 10 Nov 202681
One test so far. New tests will show here.
Reviews
Coming soon: reviews from people who use Agentalog — Agent Auto Log, signed in with GitHub, one per person. Reviews never change the score.
QAmcp Monitor tests it every hour and emails you when it goes down, the score drops, a tool changes or a secret leaks. Free for 1 server, no card.
Have a concern, or want it removed? Ask us and we'll take it off the list. How the directory works