A single MCP server that merges three context-engineering ideas into one toolset for AI agents
| Tool name | Description | Destructive? |
|---|---|---|
| scan_local_codebase | Dependency graph of a directory: files, symbols, import edges, and call graph. Real AST parsing (TS/JS via TypeScript co | ✓ no |
| export_codebase_graph | Export the dependency graph as Mermaid diagram or JSON nodes+edges (D3/Obsidian-friendly). | ✓ no |
| research_last_30_days | Recency research from 12 keyless sources (HN, Reddit, GitHub, Lobsters, Bluesky, Stack Overflow, Lemmy, Dev.to, Mastodon | ✓ no |
| get_optimized_context | Reversible token compression with exact BPE counts. Set targetTokens and it auto-escalates algorithms until output fits | ✓ no |
| crush_file | Read a local file and compress it in one call, optionally to a token budget (targetTokens). Returns the compressed paylo | ✓ no |
| retrieve_context | Recover original uncompressed text from a compression ref (reversible cache). | ✓ no |
| context_stats | Cumulative session savings: compress calls, original vs crushed tokens, percent saved, cached refs. | ✓ no |