agentleFS
mhxiemhxie/Atelier/skills/decision/SKILL.md

decision

Skill5 starsChanged 35 days agoMITView on GitHub

Lightweight decision journal with analysis proportional to the stakes.

19 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use brief, plan and 2 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AnthropicSkill

    brief

    Generate contextual briefings for legal work — daily summary, topic research, or incident response. Use when starting your day and need a scan of legal-relevant items across email, calendar, and contracts, when researching a specific legal question across internal sources, or when a developing situation (data breach, litigation threat, regulatory inquiry) needs rapid context.

    Also in 3 projects that use this

  • BlockSkill

    plan

    Draft a TDD, PRD, or lightweight plan in the conversation without committing it.

    Also in 4 projects that use this

  • OpenAISkill

    audit

    Audit or critique a product flow, journey, workflow, funnel, onboarding path, checkout path, settings path, screen, or multi-step product experience by capturing screenshots first, then reporting UX, design, and accessibility findings inline from that evidence. Use Figma only when the user explicitly asks for a board. Use when the user asks to audit, review, critique, inspect, assess, analyze, evaluate, or give feedback on a product experience.

    Also in 3 projects that use this

  • MicrosoftSkill

    handoff

    Compact the current conversation into a handoff document for another agent to pick up.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 18 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Check <paths.gtd>/decisions/ first.

    line 35
  2. Very specific:

    If a framework would expose a blind spot, select one from frameworks/cross-validation.md.

    line 42
  3. Very specific:

    When using both legs, follow protocols/agent-handoff.md → Responsibilities and Voice Legs, surface disagreement, and treat a missing direct …

    line 48
  4. Distinctive:

    Don't push for a decision.

    line 53
  5. Very specific:

    Run uv run scripts/session_log.py --type decision --duration <minutes> and fill the compact Continuity, Anomalies, and Operations sections …

    line 107

“Common” is measured: kept by at least 6 in 10 of the 18 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Decision Journal

Also reachable via /hi <natural language> (e.g., /hi should I take the offer, /hi help me decide, /hi torn between). See harness/intents.toml [intents.decision] for the row's example phrases. Both paths execute this same procedure.

Capture a choice, the reason behind it, and what should cause it to be revisited.

Trigger

User says something like:

  • "I need to decide..."
  • "Should I..."
  • "Help me think through..."
  • "I'm torn between..."

Prerequisites

  1. Reuse the current decision Repomix context artifact from $hi; for direct invocation, run uv run scripts/context_bundle.py --intent decision.

The Decision Process

Step 1: Frame the Decision

Establish the decision, real options, deadline, and binding constraints. Ask only for information that could change the recommendation.

Step 2: Search for Relevant History

Check <paths.gtd>/decisions/ first. Update an existing topic record rather than creating a sibling. Use semantic or exact search only when prior goals, constraints, or decisions could change the answer.

Step 3: Analyze Proportionally

  • Clear or reversible choice: reason directly.
  • If a framework would expose a blind spot, select one from frameworks/cross-validation.md. Add a second framework or a decision matrix only when the first pass leaves material uncertainty.
  • For a costly, irreversible, or high-uncertainty choice, dispatch the native Thinker. Add the direct Thinker leg only when independent framing could materially change the outcome.
  • When using both legs, follow protocols/agent-handoff.md → Responsibilities and Voice Legs, surface disagreement, and treat a missing direct leg as a soft downgrade.

Step 4: Decision Record

Don't push for a decision. If the user is ready, capture it. If not, capture the analysis.

Output

File: <paths.gtd>/decisions/<slugified-topic>.md

Present the proposed record or update and obtain approval before saving it. The decision itself, agreement with the analysis, and permission to write are distinct; do not treat a discussion as approval to persist it.

Slugify the topic for the stable filename: lowercase, replace spaces with hyphens, and remove special characters (e.g., "SF vs NYC job" → sf-vs-nyc-job). Dates belong in frontmatter and the decision log, never in the filename. On later sessions, update the current decision and append a dated log entry without rewriting prior entries.

---
type: decision
status: open | decided | superseded
created: YYYY-MM-DD
updated: YYYY-MM-DD
review: YYYY-MM-DD or trigger
---

## Topic

[Decision description]

## Current Decision

[Current decision and rationale, or what remains open]

## Options Considered
1. [Option A]: [description]
2. [Option B]: [description]

## Evidence and Constraints
- [Fact, assumption, constraint, or unresolved uncertainty]

## Linked Notes
- [[Note Title]] — [relevance]

## Revisit Triggers
[Date, event, or evidence that should reopen the decision]

## Decision Log

### YYYY-MM-DD
- [Decision, change, or evidence update]

Session Log

Run uv run scripts/session_log.py --type decision --duration <minutes> and fill the compact Continuity, Anomalies, and Operations sections per protocols/session-log.md. Keep the revisit trigger and actual write outcome; do not duplicate the decision analysis in the log. Warn on a logging failure without blocking the conversation.

Wrap Up

The approved stable decision file is the durable output; the compact log carries continuity. Daily notes remain read-only. Report whether the decision record was created, updated, or not saved, and its location when saved.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.