Research
Token-efficient research knowledge base with SQLite FTS5, source tier scoring, and claim verification. Persists to ~/research with project symlinks.
The Answer
A personal research knowledge base that keeps findings searchable across projects, tiered by source quality, and cheap on tokens. One ~/research/ directory, symlinks into every project that needs it.
The Problem It Solves
Research done inside a Claude Code session disappears when the session ends. The same questions get re-asked, the same sources re-fetched, contradictions across sessions never surface. Storing research as scattered markdown fixes persistence but leaves search, retrieval cost, and source trust unsolved.
How It Works
The research skill auto-activates on phrases like “investigate X” or “what’s the current state of Y”. It runs a retrieval pass against the local KB first, fills gaps from the web, then writes the merged findings back.
Storage splits content from state. RESEARCH_CONTENT_DIR holds the canonical markdown corpus plus archive, inbox, and project symlinks. RESEARCH_INDEX_DIR holds the SQLite FTS5 index, the linked-project registry, verifier logs, and extract cache. FTS5 keeps full-text search on thousands of notes in the low milliseconds.
What Makes It Different
Sources are scored against a four-tier rubric (T1 official docs and peer-reviewed through T4 forums and personal blogs), so downstream answers weigh evidence honestly. Optional symbolic verification with sympy catches math claims language models commonly fumble. Lifecycle fields mark notes as active, stale, or archived, so last quarter’s confident claim does not get cited as current.