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/reading-rivalsWrote 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/reading-rivals)<a href="https://agentmods.dev/skills/romainsimon/skills-for-decision-making/reading-rivals"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/reading-rivals/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/reading-rivals"><img src="https://agentmods.dev/badge/skills/romainsimon/skills-for-decision-making/reading-rivals.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 | $0.00086 | $0.02031 |
| Opus 5 | $0.00043 | $0.01015 |
| Sonnet 5 | $0.00017 | $0.00406 |
| Haiku 4.5 | $0.00009 | $0.00203 |
Grade A, and why
reading-rivals 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 — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reading rivals
The mistake is treating a competitor's behaviour as weather. It is not: they respond to you. A move evaluated against their current behaviour is evaluated against a world that will not exist once you make the move.
Three questions, in this order:
- What are they actually optimising? (usually not what you assume)
- What is their best response to our move?
- Where does the exchange of responses end up, and is that end state acceptable?
Question 3 is the one people skip, and it is where price wars come from.
Workflow
- [ ] 1. Infer what they optimise, from what they have done
- [ ] 2. Build the payoff table for the moves in play
- [ ] 3. Check for a dominant strategy and an equilibrium
- [ ] 4. Check whether responses cycle
- [ ] 5. Adjust for the fact that they are not perfectly rational
- [ ] 6. Decide whether to play the game or change it
1. Infer what they optimise
Do not assume they maximise profit. Observed behaviour reveals the objective, and a rival funded for growth, optimising for logo count, or run by someone who wants to sell in eighteen months will make moves that look irrational against a profit objective and are perfectly rational against theirs.
Method: list their last five or six visible moves and ask which objective makes all of
them sensible at once. See references/inferring-objectives.md. Getting this wrong makes
every subsequent step wrong, because you will be computing best responses to the wrong
payoff.
2. Build the payoff table
Two players, two to three moves each. Larger tables are not more accurate, only harder to fill in, and every cell you cannot source is a number you invented.
{
"players": ["us", "rival"],
"actions": { "us": ["hold", "cut"], "rival": ["hold", "cut"] },
"payoffs": {
"hold|hold": [100, 100], "hold|cut": [55, 130],
"cut|hold": [130, 55], "cut|cut": [70, 70]
}
}
Payoffs are annual profit contribution, same unit for both players. Estimate theirs from their pricing, their headcount, and public signals; it will be rough, and rough is enough because the structure usually determines the answer rather than the exact numbers.
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 · 198 lines · 86 tokens per session scan A 5e2568218875
reading-rivals is a skill published in the GitHub repository romainsimon/skills-for-decision-making (11 stars, last pushed 1mo ago), licensed MIT. It adds 86 tokens to every session and 2,031 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.