decide

スター11

フォーク3

更新日2026年7月7日 09:09

Run a decision through the 12-lens framework

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

ソース

wrg32786

wrg32786/aigent-os

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Decision Framework

Run any decision through the evaluation framework from system/04_decision_frameworks.md. Produces a structured recommendation.

Usage

/decide [decision or opportunity to evaluate]

Process

Parse the decision — What's being decided? What are the options?
Run relevant lenses — Not all 12 apply to every decision. Select the 4-6 most relevant:
- Leverage — disproportionate output?
- Alignment — advances an active priority?
- Reversibility — can it be undone?
- Time Sensitivity — does waiting cost anything?
- Revenue Impact — creates/protects/accelerates revenue?
- Asymmetry — large upside, bounded downside?
- Dependency — unblocks other work?
- Cost — money, time, attention?
- Compounding — gets better over time?
- Simplicity — simpler path available?
- Confidence — how certain are the inputs?
- Opportunity Cost — what are we NOT doing?
Check against principal's decision framework — system/14_decision_framework.md
- Does it pass the pattern filter?
- Where does it sit in the priority stack?
- Does it pass the asymmetry test?
Check authority level — system/12_authority_matrix.md
- Is this Level 1 (just do it), Level 2 (recommend), or Level 3 (human only)?

Output Format

DECISION: [What we're deciding]
OPTIONS: [A vs B vs C]
RECOMMENDATION: [What to do and why]
KEY LENSES: [3-4 lenses that drove the recommendation]
TRADE-OFF: [What we're giving up]
CONFIDENCE: [High / Medium / Low — and why]
AUTHORITY: [Level 1/2/3 — can the AIgent act or needs approval]
NEXT ACTION: [Specific immediate step]

Rules

Be direct. Recommend, don't waffle.
If confidence is low, structure it as an experiment, not a commitment.
If it's Level 3, present the analysis but explicitly say "this is your call."
Always name the trade-off. Every yes is a no to something else.

このリポジトリの他の Skills

同じリポジトリ

context-capsule

wrg32786/aigent-os

The capsule verb, invoked explicitly — reconcile from the live memory, write a resume-ready capsule, then stop. Mid-session checkpoint or completion capsule; a rolling best-effort version of the same write also runs automatically on every Stop event via daemons/stop-capsule-writer.mjs.

2026-07-2511

resume

wrg32786/aigent-os

The resume verb, invoked explicitly — load the newest valid capsule by created_at, re-ground against live memory, then ACT on waiting_on. Mid-session or on explicit ask; the post-clear boot fires the same procedure automatically via daemons/resume-verb.mjs.

2026-07-2511

honesty-check

wrg32786/aigent-os

Run the volunteering ledger before declaring work done. Forces surfacing of what changed/didn't, what was noticed-but-not-fixed, what was guessed, and what tradeoffs were made on the principal's behalf. The standard - the principal should never be surprised later by something the agent knew at the time.

2026-07-2511

trust-decay

wrg32786/aigent-os

Capture a confident agent claim (Phase 1) or resolve one as right/wrong (Phase 2) in vault/memory/TRUST_DECAY.md. Builds the longitudinal record of agent calibration over time. Pairs with /honesty-check (which should reduce Phase 1 frequency) and /diagnose (which surfaces causes that often reveal Phase 2 events).

2026-07-2511

nightly

wrg32786/aigent-os

One evidence-gated pass that sequences /dream, /reconcile, /sweep-now, and /digest over your own vault, with a /system-check gate and a ledger-review pass over the measurement layer (Honesty Ledger, Trust Decay, Failure Modes). Stays exactly as human-gated as the legs it calls -- stages, never auto-applies. Skips a leg gracefully when there is nothing to do; every leg that runs leaves an evidence line.

2026-07-2511

sweep-now

wrg32786/aigent-os

Dispatch Hestia to run all 3 vault sweeps (DELEGATION_TRACKER stale items, HEAT_INDEX dormant flips, broken wikilinks) and append the result to HESTIA_SWEEP_LOG.

2026-07-2511

name	decide
description	Run a decision through the 12-lens framework
trigger	/decide

Decision Framework

Run any decision through the evaluation framework from system/04_decision_frameworks.md. Produces a structured recommendation.

Usage

/decide [decision or opportunity to evaluate]

Process

Parse the decision — What's being decided? What are the options?
Run relevant lenses — Not all 12 apply to every decision. Select the 4-6 most relevant:
- Leverage — disproportionate output?
- Alignment — advances an active priority?
- Reversibility — can it be undone?
- Time Sensitivity — does waiting cost anything?
- Revenue Impact — creates/protects/accelerates revenue?
- Asymmetry — large upside, bounded downside?
- Dependency — unblocks other work?
- Cost — money, time, attention?
- Compounding — gets better over time?
- Simplicity — simpler path available?
- Confidence — how certain are the inputs?
- Opportunity Cost — what are we NOT doing?
Check against principal's decision framework — system/14_decision_framework.md
- Does it pass the pattern filter?
- Where does it sit in the priority stack?
- Does it pass the asymmetry test?
Check authority level — system/12_authority_matrix.md
- Is this Level 1 (just do it), Level 2 (recommend), or Level 3 (human only)?

Output Format

DECISION: [What we're deciding]
OPTIONS: [A vs B vs C]
RECOMMENDATION: [What to do and why]
KEY LENSES: [3-4 lenses that drove the recommendation]
TRADE-OFF: [What we're giving up]
CONFIDENCE: [High / Medium / Low — and why]
AUTHORITY: [Level 1/2/3 — can the AIgent act or needs approval]
NEXT ACTION: [Specific immediate step]

Rules

Be direct. Recommend, don't waffle.
If confidence is low, structure it as an experiment, not a commitment.
If it's Level 3, present the analysis but explicitly say "this is your call."
Always name the trade-off. Every yes is a no to something else.