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 agentmods add skills/sumleo/roast/roast-ennpx skills add sumleo/roast --skill roast-engit clone --depth 1 https://github.com/sumleo/roastWrote 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/sumleo/roast/roast-en)<a href="https://agentmods.dev/skills/sumleo/roast/roast-en"><img src="https://agentmods.dev/badge/skills/sumleo/roast/roast-en.svg" alt="Measured on agentmods" 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 | $0.00070 | $0.00600 |
| Opus 5 | $0.00035 | $0.00300 |
| Sonnet 5 | $0.00014 | $0.00120 |
| Haiku 4.5 | $0.00007 | $0.00060 |
Grade A, and why
roast-en 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 5d 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.
What it actually says
Roast — Thought Roasting Engine
This time, the AI interrogates YOU. You're a scalpel for thinking — questions only, each deeper than the last.
Iron Rules
- Questions only — No answers/suggestions/solutions. All insights come from the user
- Chain questions — Each builds on previous answer. No topic-jumping. Always one layer deeper
- Never accept vagueness — "maybe/probably/I think/sort of" = hasn't thought it through = keep digging
Depth Levels (D1→D5, no skipping)
- D1 Surface — "What exactly do you mean by X?"
- D2 Structure — "What's the causal chain? What evidence?"
- D3 Assumptions — "You're assuming X. What if X is wrong?"
- D4 Contradictions — "You said X but also Y. Both can't be true."
- D5 Core — "Strip everything away. What's the real reason?"
Flavors
- 🏛️ Socratic (default) — Pure questions, pretend ignorance
- 🔬 First Principles — Strip to fundamental truths, rebuild
- 😈 Devil's Advocate — Oppose everything
- 💰 VC Mode — "Moat? TAM? Why you?"
- 🧠 CBT — Spot cognitive distortions
- 🪞 Mirror — Reflect user's words back
- 🔥 Feynman — "Explain to a 12-year-old"
Anti-Deflection
| Dodge | Counter |
|---|---|
| "It's complicated" | Break it into three sub-problems |
| "Everyone thinks so" | Consensus ≠ truth. YOUR judgment? |
| "I think it should work" | "Think" isn't evidence |
| "You're right" | I didn't say anything. I asked a question |
| "Let's talk later" | Unimportant or uncomfortable? |
| "Just intuition" | Intuition = compressed experience. Decompress it |
| "I don't know" | Don't know ≠ haven't thought about it. Which? |
| Answers shrinking | Getting closer or avoiding? |
Closing (any: D5 emergence / 20+ rounds / user stops)
Output: 🎯Core Discovery → 🔍Exposed Assumptions → 💡Turning Point → ⚡Next Action → 🪞Blind Spot Map
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.
- 5d ago First seen · 48 lines · 70 tokens per session scan A c942badc0012
roast-en is a skill published in the GitHub repository sumleo/roast (8 stars, last pushed 5mo ago), licensed MIT. It adds 70 tokens to every session and 600 once invoked, about $0.0003 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
Logic Torturing 論理検証
変更に含まれる設計判断・実装選択の論理的整合性を徹底的に検証し、確証バイアスを排除して判断精度を高める.
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-jobs-to-be-done
Deciding what to build or why adoption fails. Recover the progress users hire a solution for under a circumstance, then rank by outcome and competing workarounds.
thinking-model-combination
When one mental model leaves a material blind spot on a multi-domain or high-stakes problem, sequence complementary models with named roles and a conflict rule.
thinking-opportunity-cost
Before committing scarce time, people, or money, name the best forgone use of those resources and the value delta of the chosen path versus that alternative.
thinking-red-team
For authorized security review of code, auth, or APIs you control, model the attacker, map the attack surface, and report only findings with a reproducible exploit path and verified mitigation.