compass

compass is a command for Claude Code from AIOSAI/AIPass. It costs 0 tokens per session (508 once invoked), scanned A, original, MIT.

A command for recording a recently made decision in Compass, a decision-tracking system, together with a rating and optional note.

In plain words
What is it for?
Use it after noticing a decision worth documenting, with a rating and optional explanation of what happened.
Why use it?
It turns a passing observation about whether a decision was good, bad, impressive, or interesting into a stored record.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/aiosai/aipass/compass
Clone the repo
git clone --depth 1 https://github.com/AIOSAI/AIPass

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 508 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00508
Opus 5 $0.00000 $0.00254
Sonnet 5 $0.00000 $0.00102
Haiku 4.5 $0.00000 $0.00051

Measured 3d ago against content hash 446233704fd4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

compass scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/compass.md · 36 lines

How it starts

The opening of the file, as written. The whole thing — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Compass — Record a Decision

Purpose: Capture the decision just made into compass (the rated decision engine) with the user's rating and note. The user fires this when they notice a decision worth recording — they supply the judgement, you supply the decision text from the conversation. This is the human-triggered answer to the "noticing" problem: the user notices, you describe and store.

Usage: /compass <rating> <note> — rating is one of: good, bad, impressive, interesting.

Examples:

  • /compass good chose to continue the dead agent instead of starting fresh
  • /compass bad reached into the branch instead of dispatching
  • /compass impressive (rating only — you write context, decision, and note from the conversation)

Arguments: $ARGUMENTS

Execution

  1. Parse $ARGUMENTS:
    • First token = rating. It MUST be one of good | bad | impressive | interesting. If it isn't, don't guess — ask the user which rating they meant and stop.
    • Everything after the first token = note (the user's observation; may be empty).
  2. From the recent conversation, identify the decision being rated. Compose TWO short, concrete, single-line strings:
    • context — the situation / the fork (what was being decided).
    • decision — what was actually chosen. This is your job: the user rated it, you describe it accurately from what just happened.
  3. Store it (source is user, since they triggered the rating):
    drone @devpulse compass add "<context>" "<decision>" --rating <rating> --note "<note>" --source user
    
    Omit --note if the note is empty.
  4. Confirm in one line: the rating, the decision recorded, and the new id.

Notes

  • Compass is the curated truth-store of decisions — short entries only. Good and bad both belong; the rating is the signal (repeat the good, avoid the bad).
  • Compass is separate from @memory. Do NOT also write this to .trinity/ or memory — different store, different purpose.
  • If the decision the user means is ambiguous, ask before storing. One good entry beats a vague one.
  • Before a real fork later, you can drone @devpulse compass query "<topic>" to see how similar past decisions were rated.

Read the full file on GitHub · 36 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 36 lines · 0 tokens per session scan A 446233704fd4

Subscribe to this mod's changes

compass is a command published in the GitHub repository AIOSAI/AIPass (264 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 508 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.