| name | dory-search |
| description | Search Dory memory before factual claims about people, projects, decisions, dates, prior sessions, or cleanup markers; follow search results with exact gets. |
Dory Search
Use dory search before answering factual questions about people, projects, decisions, dates, or prior sessions.
Workflow:
- If the task asks for a fast recent recap, use
dory_digest first instead of searching for a digest tag.
- Start normal fact retrieval with
uv run dory --corpus-root <corpus> --index-root <index> search "<query>"
- If needed, follow with
dory get <path> on the best hit
- If a result has
stale_warning, treat it as possibly outdated and look for fresher evidence
Rules:
- Prefer retrieval over guessing
- Use search even if wake seems related; wake is cached context, search is live evidence
- Do not assume digest files have a searchable
digest tag; dory_digest resolves digests/daily, digests/weekly, and legacy logs/** paths directly
- Use
mode=exact for cleanup markers, unique strings, and artifact existence checks
- Use
mode=text, mode=keyword, or mode=lexical when you want BM25-only search; all three normalize to bm25
- Use
mode=semantic when you want vector-only search; it normalizes to vector
- Hybrid search is deterministic by default; LLM-assisted planning, expansion, and reranking require explicit
DORY_QUERY_* server flags
- Use
rerank="false" for baseline deterministic search, rerank="true" to force the configured reranker, and rerank="auto" to follow server defaults
