decision-bias-check
Skill8 starsChanged 58 days agoMITView on GitHub
Audit the decision before it ships. Use for "are we sure about this?", high-stakes choices made on fast intuition, plans with no failure scenario, estimates that are suspiciously optimistic, metrics being gamed (Goodhart), unfalsifiable claims, negotiation prep, or strategies that break in extreme conditions.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- boundary-designSkill
- causal-auditSkill
- estimationSkill
- evidence-synthesisSkill
- experience-and-transmissionSkill
- failure-forensicsSkill
- formal-correctnessSkill
- measurement-disciplineSkill
- narrative-sensemakingSkill
- normative-designSkill
- problem-reframingSkill
- representation-and-possibilitySkill
- structure-discoverySkill
- systems-leverageSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 43
If no shape above matches, use a built-in agent with the closest GENIUS: pattern from rules/agent-routes.json instead.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Decision Bias Check
Problem shape: a decision, plan, or evaluation is about to be committed and nobody has run the adversarial pass: which bias produced it, what would refute it, how it behaves in the tails, and whether the metric still measures what it was meant to.
Relevant geniuses
| Agent | Use when |
| kahneman | fast intuition on a high-stakes call; no pre-mortem; estimate needs a reference class; an easier question was answered instead of the hard one |
| taleb | exposure to volatility unclassified (fragile/robust/antifragile); improvement always by addition; decision-makers carry no downside |
| popper | claim has no observation that could refute it; only easy confirmations offered as evidence; plan too big to test piecemeal |
| simon | optimizing where satisficing is rational; the search space needs decomposing before choosing |
| boyd | adversarial setting — the other side adapts; tempo and orientation matter more than the single best move |
| zhuangzi | the metric has become the target (Goodhart); the evaluation framework itself needs auditing |
| ibnalhaytham | an authority's claim taken on trust — systematic doubt with controlled isolation of variables |
| rogerfisher | multi-stakeholder deadlock — separate interests from positions, find the BATNA and the zone of agreement |
Invocation
- Pick the best-fit agent above. If two or more fit, run
tools/genius-invoker.sh route "<problem>"and take the top ranked match. - Load it:
tools/genius-invoker.sh invoke <agent> "<problem>", then readagents/genius/<agent>.mdin full. - Apply the agent's
<workflow>step by step and answer in its<output-format>. The output names the specific bias/failure with its evidence — not a generic "consider other perspectives". - Typical chain: kahneman audits the intuition → popper designs the severe
test → taleb classifies the tail exposure. Run via
tools/genius-invoker.sh compose kahneman popper -- "<problem>". - If no shape above matches, use a built-in agent with the closest
GENIUS:pattern fromrules/agent-routes.jsoninstead.
Refuse when
- The requester wants validation, not audit — a bias check that cannot change the decision is theater.
- The decision is already irreversible and the check would only assign blame.
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
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.