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/tracking-beliefsWrote 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/tracking-beliefs)<a href="https://agentmods.dev/skills/romainsimon/skills-for-decision-making/tracking-beliefs"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/tracking-beliefs/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/tracking-beliefs"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/tracking-beliefs.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 | $0.01617 |
| Opus 5 | $0.00042 | $0.00809 |
| Sonnet 5 | $0.00017 | $0.00323 |
| Haiku 4.5 | $0.00008 | $0.00162 |
Grade A, and why
tracking-beliefs 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 12d 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 — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tracking beliefs
You never observe the state of the business. You observe noisy readings of it. Revenue lags settlement, refunds and disputes arrive late, annual plans lump, analytics drops traffic to blockers and bots, and attribution is approximate everywhere.
Two operations, and almost every metric argument needs one of them:
- Filtering - given a series of noisy readings, what is the underlying level, and is the latest move larger than noise explains?
- Updating - given evidence, how should belief shift across competing explanations?
Workflow
- [ ] 1. Say what you are actually measuring, and what the reading is a proxy for
- [ ] 2. Decide whether the latest move is real
- [ ] 3. If it is real, list the explanations before looking for evidence
- [ ] 4. Update on the evidence
- [ ] 5. Act only when the belief is concentrated enough to matter
1. Reading versus state
Write both lines:
- State (unobserved): "how much recurring revenue we are actually earning."
- Reading (observed): "what Stripe reported as settled this week."
The gap between them is the noise model, and naming it usually resolves the argument before any maths. A weekly revenue reading with settlement lag and refunds has several percent of noise on it before anything real has happened.
2. Is the move real?
node scripts/calc.js track metric.json
Input shape: examples/track.json. The filter separates drift in the true level from
measurement noise and reports, per reading, whether it fell outside what drift alone
explains.
Read two things: the current level with its interval, and whether the latest reading is surprising. A reading inside the band is not evidence of anything and should not start a project. That single check kills most metric panic.
Supply processVar and observationVar when you know them. Without them the tool
guesses from the series itself and says so; the guess is a starting point, not a
measurement.
The intuition for the two numbers: observationVar is how much the reading bounces
when nothing changed - measurable by reading the same period twice, days apart, and
seeing how much it moves. processVar is how much the true level genuinely drifts per
period.
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.
- 12d ago First seen · 169 lines · 84 tokens per session scan A 3f91436f150d
tracking-beliefs 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 1,617 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.