Ref Query

What it does

Example prompts

• “what do we know about Acme Corp as of today?”
• “which engagements are active and expiring in 90 days?”
• “list our competitors”
• “/ref-query”

Requirements

• Pre-approved tools (allowed-tools): Read, Bash, Glob, Grep

Workflow steps

4 steps, taken from the step headings in SKILL.md.

1. 1Classify the question
2. 2Scan (deterministic filters)
3. 3Apply read-time interpretation (mandatory)
4. 4Answer

What it can do on your machine

Read from SKILL.md and the folder at commit fd5e9a4. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Bash
  • Glob
  • Grep

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • git

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Ref Query loads about 1.2k tokens when it runs. Until then it costs about 108 tokens; SKILL.md has 426 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~108

When it runs · the whole SKILL.md, loaded when a task matches

~1.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Bash, Glob, Grep

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Abilityai/cornelius at commit fd5e9a4, republished under its MIT licence (© Abilityai). 426 words, ~1,213 tokens.

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 Respects status (active over superseded), validity windows, and as_of freshness, and ALWAYS prints the as_of date so staleness is legible.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.

allowed-tools

Read, Bash, Glob, Grep

automation

manual

user-invocable

true

argument-hint

[scope=Company] <entity name | filter, e.g. 'engagements active expiring 90d' | 'competitors'>

metadata.version

1.0

metadata.created

2026-07-08

metadata.author

Cornelius

metadata.changelog

1.0: Initial version — deterministic frontmatter scan + fuzzy fallback with read-time interpretation (active>superseded, valid_until, as_of discount vs…

cd Brain/Company
# all active competitors
grep -rlE '^relationship: competitor' */*.md | while read f; do grep -qE '^status: active' "$f" && echo "$f"; done
# active engagements + their as_of
for f in engagements/*.md; do grep -qE '^status: active' "$f" && printf "%s  as_of=%s valid_until=%s\n" \
  "$(basename "$f" .md)" "$(grep -m1 '^as_of:' "$f" | cut -d' ' -f2)" "$(grep -m1 '^valid_until:' "$f" | cut -d' ' -f2)"; done

© Abilityai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/ref-query of Abilityai/cornelius.

Open the folder on GitHubat commit fd5e9a4