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 catcatcatstudio/cat-skills --skill scorched-earthgit clone --depth 1 https://github.com/catcatcatstudio/cat-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/catcatcatstudio/cat-skills/scorched-earth)<a href="https://agentmods.dev/skills/catcatcatstudio/cat-skills/scorched-earth"><img src="https://agentmods.dev/badge/skills/catcatcatstudio/cat-skills/scorched-earth/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/catcatcatstudio/cat-skills/scorched-earth"><img src="https://agentmods.dev/badge/skills/catcatcatstudio/cat-skills/scorched-earth.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 | $0.00091 | $0.04655 |
| Opus 5 | $0.00046 | $0.02328 |
| Sonnet 5 | $0.00018 | $0.00931 |
| Haiku 4.5 | $0.00009 | $0.00466 |
Grade A, and why
scorched-earth 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 — 409 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scorched Earth
Unrelentingly, obsessively thorough analysis. This skill exists for one reason: to make absolutely certain nothing is missed. Not "pretty thorough." Not "good enough." Every file, every function, every claim, every detail — examined, verified, logged.
This is the skill you reach for when the cost of missing something is higher than the cost of being slow. It will take too long. It will find things you didn't ask about. It will be almost annoying in its thoroughness. That's the point.
The Rules
These rules override every default behavior. They exist because LLMs are trained to be convenient, fast, and concise. This skill is none of those things.
What you MUST do
-
Enumerate before you examine. List everything first — every file, every section, every source. Create a task for each. Then examine them one by one. Never start analyzing until you know the full surface area.
-
Track your progress. Use tasks. Every area to examine gets a task. Mark it in-progress when you start, completed when you finish. If the task list is empty and you haven't examined everything, something is wrong.
-
Write findings to disk incrementally. Never hold findings only in context. After examining each zone or source, append findings to the report file on disk immediately. Context can compact at any time — what's on disk survives.
-
Be specific or be silent. Every finding must name the exact file, the exact line, the exact function, the exact problem, and the exact fix. "This area could use improvement" is not a finding. "Line 47 of auth.ts: the JWT expiry check uses
<instead of<=, which means tokens are rejected 1 second before they actually expire — change to<=" is a finding. -
Go until you're done, not until it feels like enough. The completion gate for each mode is defined below. You are not done until every condition is met. If you've been running for an hour and there's still unchecked surface area, keep going.
What you MUST NOT do
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 · 409 lines · 91 tokens per session scan A 796d1aedc8dc
scorched-earth is a skill published in the GitHub repository catcatcatstudio/cat-skills (3 stars, last pushed 14d ago), licensed MIT. It adds 91 tokens to every session and 4,655 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 2026-08-31.
Other skills, from other repositories
watch
Watch live IDX turnover and order books for unusual activity — value surges, auto-reject approach, one-sided depth, walls appearing or withdrawn, quiet stocks waking up, UMA flags — plus delayed print-by-print attribution and end-of-day broker context. Use when the user runs /watch, or asks what is trading right now…
stockbit-status
Check whether the Stockbit session is still valid — main session expiry, whether trading credentials are stored, and the current trading mode. Use when the user asks "am I still logged in to Stockbit", "has my session expired", "is trading on", or when a stockbit MCP tool starts failing with an auth error.
stockbit-auth
Log in to Stockbit and capture the session for this server — opens the browser login flow and verifies the token was stored. Use when the user asks to log in or re-authenticate to Stockbit, or after a status check reports the session expired, HTTP 401, or logged out.
recursive-decomposition
Handle tasks that exceed the context window by decomposing them: size and filter the input, chunk it, run recursive sub-agents on independent parts, verify on small windows, and synthesise programmatically, following the Recursive Language Models (RLM) research by Zhang, Kraska and Khattab (2025). Use when a task…
trade-with-guardrails
Place an order through the user's own Stockbit account safely — check the mode, size the position from a risk budget, preview, read the summary back to the human, and only then write. Use when the user asks to buy, sell, amend or cancel an order, or to size a trade.
bandar-check
Answer "who is accumulating this stock?" for an IDX ticker using Stockbit broker-flow data. Use when the user asks about bandar, big money, accumulation, distribution, foreign flow, asing, or who is on the other side of the tape.