cachly AI Brain
Description
Persistent AI memory for Claude Code, Cursor, Copilot, Windsurf, Cline & Zed. Your AI remembers every lesson, fix, and architecture decision — across sessions, IDEs, and model upgrades. Quality-aware recall beats raw BM25 keyword search by **+33% Precision@1** on our published benchmark. Features causal root-cause analysis (causal_trace), failure prediction (brain_predict), git-ambient learning (brain_from_git), shared Team Brain, and Memory Crystals. BM25+ search in 11 languages. **122 tools. Free tier. GDPR, EU-hosted on German servers.** Zero-config: sign in once at [cachly.dev/setup-ai](https://cachly.dev/setup-ai) — your Brain instance is provisioned automatically. No credit card required.
What our measurements say
cachly AI Brain is a remote MCP server published at mcp.cachly.dev. It has been probed 7 times since 9/10/2026. It answered in 7 of them (100.0%), a near-uninterrupted record. Median response time is 286 ms, placing it among the faster endpoints. It exposes a broad tool surface of 27 tools. On the protocol side it still runs 2025-11-25 and has not moved to the newer spec.
Agent readability score
Can an LLM agent pick the right tool here — names, descriptions and parameter clarity are assessed.
Agent readability score
Strengths
- Tool names like 'cache_get' and 'cache_set' clearly indicate read/write operations
- Descriptions often specify use cases (e.g., 'index_project' for file indexing)
- Some tools (e.g., 'brain_doctor') provide actionable health diagnostics
Weaknesses
- Ambiguous tool names like 'semantic_search' vs 'smart_recall' with overlapping purposes
- Technical jargon in descriptions (e.g., 'CKG nodes' without explanation)
- Incomplete parameter documentation (e.g., 'grund' parameter in 'learn_from_attempts' lacks definition)
semantic_search— Overlaps with smart_recall in semantic search functionalitybrain_predict— References CKG/Redis without defining acronymslearn_from_attempts— Unclear 'grund' parameter definitioncausal_trace— No explicit error handling for missing prior learningcachly_tool— Generic name with no specific use case clarification
Verdict
Event narrative
Risk: medium
- 2026-09-10T21:16:35.016Z — era changed from 'auth-gated' to 'legacy', complianceScore increased from null to 100 · indicates successful transition to legacy mode with full compliance
Advice
Schema compliance
tools/list structure, inputSchema validity, and a functional smoke test — the components of the 0-100 score.
- tools/list schema valid
- inputSchema is valid JSON Schema
- Smoke test passed · last run:
- Cold-start: cold 32ms → warm 30ms (penalty 2ms)
Availability history
Event log
Derived by comparing consecutive probes — changes in era, protocol version, build and reachability.
- Server build: — → 0.10.169
- Supported versions: — → 2025-11-25
- Era changed: Auth-gated → Legacy
- Added to watch — era: Auth-gated
Get your badge
Add this badge to your README — it updates automatically as measurements change.
[](https://mcpmetrics.io/servers/cachly-dev-cachly-mcp)<a href="https://mcpmetrics.io/servers/cachly-dev-cachly-mcp"><img src="https://mcpmetrics.io/badge/cachly-dev/cachly-mcp/era.svg" alt="mcpmetrics"></a>You are seeing the last 7 days. Sign up for the full history. Which check failed and why is in the dashboard.
Sign up free to see