ref-query

name	ref-query
description	Structured, temporal lookup against a reference scope — "what do we know about Acme Corp as of today?", "which engagements are active and expiring in 90 days?", "list our competitors". Respects status (active over superseded), validity windows, and as_of freshness, and ALWAYS prints the as_of date so staleness is legible. Distinct from /recall, which searches cognitive insights. Company is the default scope. Read-only.
automation	manual
allowed-tools	["Read","Bash","Glob","Grep"]
user-invocable	true
argument-hint	[scope=Company] <entity name | filter, e.g. 'engagements active expiring 90d' | 'competitors'>
metadata	{"version":"1.0","created":"2026-07-08T00:00:00.000Z","author":"Cornelius","changelog":["1.0: Initial version — deterministic frontmatter scan + fuzzy fallback with read-time interpretation (active>superseded, valid_until, as_of discount vs per-type SLA, always print as_of)"]}

Ref Query

ℹ️ First, set expectations: before anything else, print one short line with this skill's version and its most recent change — the top entry of metadata.changelog above — e.g. ref-query v1.0 — recent: initial version. Then proceed.

Answer a structured question about reference records with the temporal contract applied. This is not /recall — /recall searches the user's cognitive insights; ref-query reads the CRM/market source of truth and honours status/validity/as_of.

Read first: resources/layered-brains/COMPANY-BRAIN-SCHEMA.md → "The temporal model" (read-time interpretation + the per-type Freshness SLA). This skill enforces those rules.

State Dependencies

Source	Location	Read	Write	Description
Company family	Brain/Company/{people,orgs,products,engagements,market}/*.md	✓		Entity notes + frontmatter
Search wrapper	resources/local-brain-search/run_search.sh	✓		Fuzzy "tell me about X" under a core,Company mount

Process

Step 1 — Classify the question

• Entity lookup ("what do we know about X as of today?") → resolve the note (Grep by name/alias, or a mounted search) and read it.
• Filtered list (by type/relationship/status/expiry window) → deterministic frontmatter scan (below), not semantic search — the answer must be exact and reproducible.
• Fuzzy / exploratory ("who might overlap with X?") → mounted semantic search:

  BRAIN_READ_SCOPE=core,Company resources/local-brain-search/run_search.sh "<query>" --limit 10 --json
  

Step 2 — Scan (deterministic filters)

Frontmatter is the query surface. Examples:

cd Brain/Company

grep -rlE  */*.md |   f;  grep -qE   &&  ; 

 f  engagements/*.md;  grep -qE   &&   \
    ;