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/valuing-informationWrote 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/valuing-information)<a href="https://agentmods.dev/skills/romainsimon/skills-for-decision-making/valuing-information"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/valuing-information/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/valuing-information"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/valuing-information.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 | $0.00084 | $0.02224 |
| Opus 5 | $0.00042 | $0.01112 |
| Sonnet 5 | $0.00017 | $0.00445 |
| Haiku 4.5 | $0.00008 | $0.00222 |
Grade A, and why
valuing-information 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 — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Valuing information
The value of an observation is the increase in expected utility from making it:
VOI(O) = sum over outcomes P(o) * max_a EU(a | o) - max_a EU(a)
It is never negative. It is exactly zero whenever the best action is the same whatever you learn. That zero is the entire point of this skill: most proposed research is decoration, and this is the calculation that says so before the money is spent rather than after.
Workflow
- [ ] 1. Name the decision the information would serve
- [ ] 2. State the actions and what you would do today with no new data
- [ ] 3. State the hidden variable, its states, and the prior
- [ ] 4. Score every action-state pair
- [ ] 5. Compute the perfect-information ceiling before pricing any real study
- [ ] 6. State honestly how discriminating each candidate observation is
- [ ] 7. Run the calculator and read the "changes decision" column
- [ ] 8. If it is worth it, size the study
Put the tables in the answer. The decision table and the likelihood table are not scratch work: write them out. A reader who cannot see the numbers cannot challenge them, and the numbers are the only part of this that is falsifiable.
1. Name the decision
Write the sentence: "We are trying to decide whether to ___, and we would act differently if we learned ___."
If you cannot complete it, there is no decision, so there is no value of information. That does not make the research illegitimate - curiosity and background understanding are real - but it does mean it should compete for time as a discretionary activity, not be waved through as diligence.
2-4. Build the decision
Same construction as the framing-decisions skill: actions, states, prior, one utility scale. If that skill has already been run, reuse its table verbatim.
Write the table out, states as rows and actions as columns:
| State | P | raise 40% | raise 15% | hold |
|---|---|---|---|---|
| tolerates +40% | 0.35 | 45000 | 20000 | 0 |
| tolerates +15% | 0.40 | 5000 | 18000 | 0 |
| tolerates nothing | 0.25 | -28000 | -6000 | 0 |
What ships with it
5 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 · 206 lines · 84 tokens per session scan A 31f1a2c8340c
valuing-information is a skill published in the GitHub repository romainsimon/skills-for-decision-making (11 stars, last pushed 1mo ago), licensed MIT. It adds 84 tokens to every session and 2,224 once invoked, about $0.0004 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.