Best for
- Researchers and developers who need to track and compare results across AI experiments
A
AI Experiment Logger
Community
Track and analyze AI experiments with MCP tools and web dashboard
- Category
- Dev Tools #433 of 438
- Stars
- ★ 2.8k Very popular
- Transport
- stdio (local process)
- Runtime
- Node.js
- Credentials
- No credential needed
- License
- MIT
- Last commit
20FMRS · D
The server has a clear purpose for AI experiment tracking, but the specific tools it provides are not listed in the source. It comes from a community-driven plugin repository, so users should evaluate its production and security readiness.
Strongest · Maintenance 5/20 Weakest · Documentation 2/20
Why each score
- Reliability 4/20
- The only evidence is the server.json manifest and the repository-wide README; no source code, tests, CI workflow, or declared tool list for this server is supplied, so a clean stdio startup and MCP init/tool-list handshake cannot be verified. The manifest is well-formed and the npm/stdio packaging makes a happy path plausible, but missing executable evidence forces a deduction.
- Security and permissions 5/20
- No red-line issue is visible (no malware, credential theft, embedded secrets, or covert exfiltration), but the absence of source makes credential handling, data boundaries, web-dashboard side effects, and external network behavior impossible to audit. Least-privilege and confirmation mechanisms are unverified and data flow is undisclosed, so the score stays low.
- Maintenance 5/20
- The parent repository shows ecosystem-level activity such as release badges and a CLI, but that is not per-server maintenance evidence. For ai-experiment-logger only a 1.0.0 manifest is available; there is no verifiable commit history, issue response, dependency update record, or server-specific changelog. The MIT license is repo-level and does not justify a higher score.
- Documentation 2/20
- Documentation consists of one catalog line in the README plus generic CLI examples. There are no server-specific installation steps, authentication details, tool parameters, limits, cost guidance, troubleshooting notes, or examples, so capabilities and limitations cannot be reviewed against actual behavior.
- Setup experience 4/20
- The manifest provides an npm package identifier and stdio transport, making an install path superficially clear. However, there are no client configuration examples, prerequisite or environment-variable documentation, platform compatibility notes, or committed tests/CI proving the setup works. Under the static calibration the score must remain low.
Static review · not runListed 2026-08-07
Read the FMRS scoring method →Fit and risk
Not for
- Users needing broad database tooling
- Users who expect a different client configuration beyond the npm package
Required permissions
- File system access to store experiment logs
- Network access to serve the web dashboard
Risks and side effects
- Experiment logs may contain sensitive data
- The web dashboard might expose unsecured endpoints
Setup
Before you start
Runtime：Node.js
Install the server: pnpm add -g @intentsolutionsio/ccpi && ccpi install ai-experiment-logger. Alternatively, add the repository as a Claude Code plugin marketplace: /plugin marketplace add jeremylongshore/claude-code-plugins, then /plugin install ai-experiment-logger@claude-code-plugins-plus.
Check that it works
After installing the plugin, confirm the AI Experiment Logger tools appear in your MCP client's tool list; since no tool names are recorded in the source, logging a test experiment is a practical connection check.
Troubleshooting
- Ensure the npm package is installed correctly
- Check that the server starts via stdio
- Verify the web dashboard is accessible
Things to try
Once connected, you can ask your AI assistant things like:
- Log the results of my latest AI experiment and show the trend
- Open the experiment dashboard and summarize recent experiments
Use cases
Track AI experiments
Analyze experiment results
Visualize experiment data
Overview
The AI Experiment Logger is an MCP server that provides MCP tools and a web dashboard to track and analyze AI experiments. It is part of the Claude Code Plugins Plus Skills repository, which contains 471 plugins, 3,069 skills, and 347 agents. The server is available as an npm package '@claude-plugins/ai-experiment-logger' and uses stdio transport. This file describes only this server; it does not cover other servers in the repository.
Similar servers
Context7
80 · B
Upstash's official server providing up-to-date third-party library docs for AI coding assistants
NPM Sentinel MCP Server
79 · B
AI-powered NPM package analysis MCP server
MartinLoop MCP Server
79 · B
An independent control layer that gives AI coding agents budgets, brakes, and receipts
GitHub MCP Server
76 · B
GitHub's official MCP server for managing repos, issues, PRs, and workflows via natural language
More servers from this repository
Source revision ebea1067983c Data synced 2026-10-12 Read the FMRS scoring method