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/stress-testing-plansWrote 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/stress-testing-plans)<a href="https://agentmods.dev/skills/romainsimon/skills-for-decision-making/stress-testing-plans"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/stress-testing-plans/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/stress-testing-plans"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/stress-testing-plans.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 | $0.00088 | $0.01851 |
| Opus 5 | $0.00044 | $0.00925 |
| Sonnet 5 | $0.00018 | $0.00370 |
| Haiku 4.5 | $0.00009 | $0.00185 |
Grade A, and why
stress-testing-plans 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.
Stress testing plans
A plan is optimized against a model of the world. The model is wrong. Validation asks three questions, in this order:
- Does the ranking survive plausible errors in the model?
- What is the most likely way this fails, and does that trajectory worry us?
- What are we actually trading, and have we said so?
The governing principle: plan with a simple model, evaluate with a rich one. A simple planning model resists overfitting to assumptions you cannot verify. The evaluation model can be as elaborate as you can justify, because nothing is being fitted to it.
Workflow
- [ ] 1. List the assumptions the plan depends on
- [ ] 2. Score each option under each plausible variant
- [ ] 3. Check whether the ranking survives
- [ ] 4. Find the most likely failure and read it back
- [ ] 5. Make the trade-off explicit
- [ ] 6. Pick a response from the fixed menu
1. List the assumptions
Not risks. Assumptions: quantities the plan's ranking depends on and which you have not measured. Churn stays flat. The infra bill scales linearly. Conversion holds at the new price. No rival ships a free tier.
Keep the ones where a plausible alternative value is materially different. Three to five is the working range; more and the matrix becomes unreadable.
2-3. Does the ranking survive?
Build a matrix of option by scenario and run:
node scripts/calc.js robust plan.json
Input shape: examples/robust.json. It reports three winners:
- best expected - highest weighted average across scenarios
- best worst case (maximin) - best outcome under its own worst scenario
- least regret (minimax regret) - smallest gap to whatever turns out to be best
If all three agree, the decision is robust. Ship it and stop analysing. That is the common outcome and it is worth reaching quickly.
If they disagree, the disagreement is the decision, and it is a question about appetite rather than about arithmetic. Say out loud how much downside you are buying insurance against, decide, and record the reason. Do not let the disagreement stay hidden behind whichever number someone quoted.
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 · 185 lines · 88 tokens per session scan A d764028551c7
stress-testing-plans is a skill published in the GitHub repository romainsimon/skills-for-decision-making (11 stars, last pushed 1mo ago), licensed MIT. It adds 88 tokens to every session and 1,851 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.