Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add product-on-purpose/thinking-framework-skills --skill think-five-whysgit clone --depth 1 https://github.com/product-on-purpose/thinking-framework-skillsWrote 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/product-on-purpose/thinking-framework-skills/think-five-whys)<a href="https://agentmods.dev/skills/product-on-purpose/thinking-framework-skills/think-five-whys"><img src="https://agentmods.dev/badge/skills/product-on-purpose/thinking-framework-skills/think-five-whys/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/product-on-purpose/thinking-framework-skills/think-five-whys"><img src="https://agentmods.dev/badge/skills/product-on-purpose/thinking-framework-skills/think-five-whys.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 | $0.00094 | $0.01686 |
| Opus 5 | $0.00047 | $0.00843 |
| Sonnet 5 | $0.00019 | $0.00337 |
| Haiku 4.5 | $0.00009 | $0.00169 |
Grade A, and why
think-five-whys 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Five Whys
Five Whys traces a problem to a root cause by asking "why?" of each answer, about five times, until you reach a cause worth fixing. It is famous and frequently requested, and its evidence is weak: the critical literature finds it flattens multi-cause problems into a single chain and discards the rest. This skill runs Five Whys honestly. It leads with that caveat, then forces the one discipline a bare why-chain omits and that makes it safe to run at all: at each step, flag whether more than one cause could apply, and the moment the problem is multi-cause, say so and hand it to think-issue-tree. The output is a single linear why-chain with a branch-or-not flag on each step plus one countermeasure aimed at the terminal node, not a confident single root cause presented as the cause.
Before you run this: the evidence caveat
Five Whys is tier X (weak or contradictory evidence). The governing critique is Alan J. Card, "The problem with '5 whys'," BMJ Quality & Safety 26(8) (2017): 671-677, which argues the technique oversimplifies complex problems, follows a single causal chain when failures are multi-causal, and limits understanding of how processes actually fail; Card's prescription is to move to branching, systems-oriented causal-tree diagramming. John Allspaw, "The Infinite Hows (or, the Dangers of the Five Whys)" (2014), adds that in socio-technical systems the linear "why" chain drifts toward a person and a counterfactual (the "Bad Apple" blame trap) and recommends "how" over "why" to recover the multi-contributor picture. The method is reliable only for simple, linear, single-cause failures; beyond them its output looks like THE cause when it is one cause among many.
So a single chain is not a verdict. If you run Five Whys here, the value is not the chain, it is the discipline this skill adds: flagging at every step whether more than one cause could apply, and stopping to redirect when it can. For any problem that might have more than one cause - which is most real problems - think-issue-tree is the stronger move: it branches with MECE-plus-remainder discipline instead of forcing one path. Do not report a single root cause as the answer to a many-cause problem, and do not invent quantified claims about Five Whys' effectiveness (none are supported).
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 · 72 lines · 94 tokens per session scan A 033ee19ef2d6
think-five-whys is a skill published in the GitHub repository product-on-purpose/thinking-framework-skills (15 stars, last pushed today), licensed Apache-2.0. It adds 94 tokens to every session and 1,686 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
cocoreview
CocoReview — structured code review with six-severity findings vocabulary, progressive disclosure architecture, and universal anti-pattern baseline. Invoked via $review [file] [--complexity] [--security] [--architecture] [--language ].
meter-reconcile
Skill-native CocoMeter transcript reconciliation and adapter canary. Replaces the former meter-reconcile and adapter-self-test runtime helpers.
afrexai-performance-engineering
Complete performance engineering system — profiling, optimization, load testing, capacity planning, and performance culture. Use when diagnosing slow applications, optimizing code/queries/infrastructure, load testing before launch, planning capacity, or building performance into CI/CD. Covers Node.js, Python, Go…
agent-anti-false-completion
A set of working rules for coding agents that emphasizes checking results, investigating related problems, and reporting honestly.
agent-bom-troubleshoot
Diagnose issues, check prerequisites, and validate configurations. Use when: "doctor", "debug", "why failing", "validate config", "check prerequisites", "something is broken", "db status", "fix my setup".
stall
Show CocoStall loop and no-progress detection status.