Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/romainsimon/skills-for-decision-makingnpx agentmods add skills/romainsimon/skills-for-decision-making/framing-decisionsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/romainsimon/skills-for-decision-making/framing-decisions)<a href="https://agentmods.dev/skills/romainsimon/skills-for-decision-making/framing-decisions"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/framing-decisions/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/romainsimon/skills-for-decision-making/framing-decisions"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/framing-decisions.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00095 | $0.01827 |
| Opus 5 | $0.00048 | $0.00914 |
| Sonnet 5 | $0.00019 | $0.00365 |
| Haiku 4.5 | $0.00010 | $0.00183 |
Grade A, and why
framing-decisions 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Framing decisions
Most stalled decisions are not hard. They are unframed: the options are fuzzy, the disagreement is about an unnamed hidden variable, and the outcomes are being scored on two different scales by two different people. Framing dissolves more arguments than analysis does.
Run the whole workflow. Skip to the calculator and you will optimize a badly posed problem, which is worse than arguing.
Workflow
- [ ] 1. Name the actions (at least three, one of which is do nothing)
- [ ] 2. Name the hidden variable the disagreement is actually about
- [ ] 3. Put a number on the prior, sourced not guessed
- [ ] 4. Score every action-state pair on one scale
- [ ] 5. Run the calculator
- [ ] 6. Run the bias check before committing
- [ ] 7. Write the decision record
1. Name the actions
List them as things someone could start doing tomorrow. Three rules:
- Always include "do nothing". It is a real action with a real payoff, and it is often the incumbent that everything else must beat.
- Look for the reversible half. If an option can be split into a cheap reversible part and an expensive irreversible part, those are two different actions. The reversible half usually wins on expected utility and is usually not on the list.
- Stop at five. More options do not improve the decision; they hide it.
2. Name the hidden variable
Ask: what would we need to know for this to be obvious? That is the hidden state. Write it as two to four mutually exclusive, collectively exhaustive states.
If two people disagree about the action but agree on the state, the disagreement is about values, and no analysis will settle it. Say so and escalate to whoever owns the values.
3. Put a number on the prior
Probabilities come from a reference class, not from a feeling. In order of preference:
- Our own past attempts at the same thing. Count them.
- A published base rate for this class of thing.
- A stated pseudocount: "this is a Beta(2,3) belief, worth about five observations."
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 185 lines · 95 tokens per session scan A 352ddd0debce
framing-decisions is a skill published in the GitHub repository romainsimon/skills-for-decision-making (11 stars, last pushed 1mo ago), licensed MIT. It adds 95 tokens to every session and 1,827 once invoked, about $0.0005 per session on Opus 5. 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 2026-08-31.
Other skills, from other repositories
thinking-model-router
When unsure which thinking skill fits, map domain and problem type, then return NONE or one primary skill by default (at most three complementary).
thinking-systems
When behavior is emergent across components—fixes elsewhere break, loops/delays dominate—map boundary, stocks/flows, feedback, archetypes, then rank leverage.
architecture-aware-init
Selects architecture paradigm via research before scaffolding. Use when architecture is undecided and the choice needs justification and documentation.
thinking-five-whys-plus
When a fault is localized and the proximate cause is known but the systemic root is not, chain evidence-linked whys with a counterfactual stop and a countermeasure.
thinking-map-territory
When a claim, doc, test, metric, or assumption conflicts with observed behavior, stop theorizing from the map and verify the live code or data; let territory overrule.
thinking-theory-of-constraints
When throughput or latency is pipeline-limited, identify the single binding constraint and exploit, subordinate, elevate, then recheck—ignore non-constraints.