An observational memory server that uses LLM agents to capture, compress, and recall project-specific decisions and context across AI coding sessions. It stores structured observations in SQLite to maintain long-term session continuity and architectural awareness.
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find.
Install from
M8ven verifies MCPs across every public registry — install directly from whichever one you prefer.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
simple-git has blockUnsafeOperationsPlugin bypass via case-insensitive protocol.allow config key enables RCE
When Vitest UI server is listening, arbitrary file can be read and executed
Vitest allows Remote Code Execution when accessing a malicious website while Vitest API server is listening
simple-git is vulnerable to Remote Code Execution
simple-git Affected by Command Execution via Option-Parsing Bypass
process.env. You'll be asked to provide them before it can run.CODEWATCH_AUTO_REFLECT— true Enable auto-reflectionCODEWATCH_CHARS_PER_TOKENCODEWATCH_DATA_DIR— ~/mcp-data/codewatch-memory/ SQLite storage locationCODEWATCH_DEFAULT_BRANCHCODEWATCH_FALLBACK_PROVIDER— openai Fallback LLM (google, openai, groq, none)CODEWATCH_GOOGLE_MODEL— gemini-2.5-flash Google modelCODEWATCH_GROQ_MODEL— llama-3.3-70b-versatile Groq modelCODEWATCH_LLM_PROVIDER— "": "groq"CODEWATCH_LOG_LEVEL— info Logging verbosityCODEWATCH_MAX_COMPRESSION— 3 Max compression level (0-3)CODEWATCH_OPENAI_MODEL— gpt-4o-mini OpenAI modelCODEWATCH_PROJECT_DIRCODEWATCH_REFLECT_THRESHOLD— 40000 Auto-reflect trigger (tokens)[](https://m8ven.ai/mcp/klausandrade-codewatch-ej0s2f)