Skip to content
Back to skills

Starlight Decision Ledger

ASecurity

Preserve a consequential venture decision, its tradeoffs, owner, review date, and supporting evidence when the user asks to decide, approve, reject, document, or audit a decision.

  • 10 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 20, 2026
ai-agents

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 20, 2026

npx -y skills add frankxai/Starlight-Intelligence-System --skill starlight-decision-ledger --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Starlight Decision Ledger?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Starlight Decision Ledger
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/frankxai-starlight-decision-ledger/badge)](https://www.skillsdirectory.com/skills/frankxai-starlight-decision-ledger)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: starlight-decision-ledger
description: Preserve a consequential venture decision, its tradeoffs, owner, review date, and supporting evidence when the user asks to decide, approve, reject, document, or audit a decision.
---

# Starlight Decision Ledger

Record the decision as a falsifiable operating commitment, not a meeting note.

1. Search or fetch existing records when the user may be revisiting an earlier decision.
2. Register source material with `register_evidence` when a URL, artifact, observation, or metric materially supports the decision. Separate sourced claims from the user's judgment.
3. Call `record_decision` with context, the actual choice, rejected tradeoffs, one owner, and a review date when reversibility matters.
4. `approved` and `rejected` are consequential statuses. Use them only after explicit user confirmation and pass `user_confirmed: true`; otherwise record the decision as `proposed`.

Do not invent consensus, evidence, approval, or an accountable owner. Return the decision ID and the evidence IDs that make the decision auditable.

Files in this skill

  • SKILL.md1.1 KB
  • agents/openai.yaml213 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…