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 deciqAI/knowledge-skills --skill hanlons-razorgit clone --depth 1 https://github.com/deciqAI/knowledge-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/deciqai/knowledge-skills/hanlons-razor)<a href="https://agentmods.dev/skills/deciqai/knowledge-skills/hanlons-razor"><img src="https://agentmods.dev/badge/skills/deciqai/knowledge-skills/hanlons-razor/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/deciqai/knowledge-skills/hanlons-razor"><img src="https://agentmods.dev/badge/skills/deciqai/knowledge-skills/hanlons-razor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00110 | $0.01618 |
| Opus 5 | $0.00055 | $0.00809 |
| Sonnet 5 | $0.00022 | $0.00324 |
| Haiku 4.5 | $0.00011 | $0.00162 |
Grade A, and why
hanlons-razor 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 9d 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hanlon's Razor
Overview
Before assuming someone hurt you on purpose, construct the version where they made a mistake — and see how much evidence it explains. The razor is a Bayesian prior, not a proof; override it when concrete evidence of malice arrives. Human attribution systematically over-weights intent (fundamental attribution error); most hostile-seeming acts are incompetence, miscommunication, or asymmetric information.
Composes with bayesian-reasoning, abductive-reasoning, occams-razor, critical-thinking.
When to Use
- You feel an emotional pull toward "they did this on purpose"
- You're about to escalate on the assumption of malice
- A pattern of bad outcomes is being framed as a coordinated attack
- A team is in conflict and each side believes the other is acting in bad faith
- An AI model's harmful/biased output or a competitor's surprising AI move is being read as deliberate malice rather than an emergent bug, honest error, or ordinary self-interested competition
Not when: concrete evidence of malicious intent exists; cost of being wrong is catastrophic; power imbalance makes "they probably didn't mean it" an abuse-enabling stance.
Coaching Novices (Adaptive Front Door)
- Engine mode: user has a concrete case → run The Process directly.
- Coach mode: user is unfamiliar → guide step by step.
In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.
- One-line: before believing someone did it on purpose, construct the mistake version — see how much evidence it covers.
- Check fit: concrete malice evidence / catastrophic cost of being wrong → not this lens.
- Elicit the specific incident — what exactly happened?
[WAIT — do not advance until user responds]
- Work through The Process one step at a time with their input.
[WAIT — do not advance until user responds]
- Close: name the clarifying-conversation move + the override signal to watch for.
[WAIT — do not advance until user responds]
What ships with it
3 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.
- 9d ago First seen · 124 lines · 110 tokens per session scan A bcb414c96634
hanlons-razor is a skill published in the GitHub repository deciqAI/knowledge-skills (10 stars, last pushed 10d ago), licensed MIT. It adds 110 tokens to every session and 1,618 once invoked, about $0.0006 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 2026-09-03.
Other skills, from other repositories
voice-builder
Ingests 5–20 of the user's writing samples and distils a reusable voice-profile.md — signature phrases, sentence-length distribution, opener/closer habits, punctuation quirks, vocabulary do/don't lists, tone sliders, and three calibration paragraphs with self-checks. Other skills load this file so every post sounds…
english-swe-daily
Daily English expression coach for intermediate software engineers. Use this skill whenever the user wants to improve their spoken or written English in a software engineering work context — especially for standups, Slack messages, 1:1s, meetings, giving feedback on code, asking for help, disagreeing politely, or…
agent-teacher
A teaching aid that explains technical ideas with a small runnable code example and a guided walkthrough. It is meant for learning how something works, not fixing existing code.
lecture-notes
Transform raw lecture content (transcript, plain text, or slides) into clean, structured, study-ready notes in Markdown. Use this skill whenever the user pastes lecture material — a transcript (with or without timestamps), course/lecture text, or slide content — and asks for notes, a summary, or help studying. Also…
skill-creation-walkthrough
Step-by-step guide for creating your own Claude Skills, from deciding whether a skill is the right tool to writing the SKILL.md file, structuring reference material, and making it trigger reliably. Use when you want to package a workflow, framework, or repeated task into a reusable Skill, when an existing skill is not…
master-debate
A skill for running an adversarial, multi-round debate between two Buddhist teachers. It is designed for questions where the teachers argue from different positions.