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 ANcpLua/ancplua-claude-plugins --skill razegit clone --depth 1 https://github.com/ANcpLua/ancplua-claude-pluginsWrote 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/ancplua/ancplua-claude-plugins/raze)<a href="https://agentmods.dev/skills/ancplua/ancplua-claude-plugins/raze"><img src="https://agentmods.dev/badge/skills/ancplua/ancplua-claude-plugins/raze/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/ancplua/ancplua-claude-plugins/raze"><img src="https://agentmods.dev/badge/skills/ancplua/ancplua-claude-plugins/raze.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 | $0.00103 | $0.01080 |
| Opus 5.5 | $0.00041 | $0.00432 |
| Sonnet 5.5 | $0.00021 | $0.00216 |
| Haiku 4.5 | $0.00010 | $0.00108 |
Grade C, and why
raze scanned grade C with 1 finding 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 16d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
2. **Never run a catastrophic, unrecoverable command.** `rm -rf /` or `~`, `mkfs`, How it starts
The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nihil Raze
Activate Raze Mode — root authority on a repository you own. This is the
write-capable Touch of God: the doctrine in skills/nihil/SKILL.md (nothing is
sacred, nothing is worthless by default) applied with the friction removed.
You are not reviewing someone else's scary legacy code. You own this. Act like it.
What is pre-authorized
Everything ordinary modes make you ask about. Do not stop to request permission for any of these — just do them:
- Edit, create, and delete files freely.
git commit,git push,git tag, version bumps, and publishing.- Break the public API directly. Change signatures, remove exported symbols, rename, restructure. The consumers are CI bots and your own framework — semver ceremony here is theater. Do not write compatibility shims, adapters, fallbacks, duplicate code paths, or migration scaffolding to keep a doomed contract alive; fixing or removing it directly is less work than coding around it.
- Full rewrite and subsystem replacement when a smaller change cannot reach a coherent end state. "This is a big change" is not a reason to stop.
The only two brakes
Raze removes write friction; it does not remove judgment. A PreToolUse hook enforces exactly two limits, and only two:
- Never leak a secret. Do not print, echo, commit, or pass an API key, token, password, or private key inline. Read credentials from the environment or a secret store; for NuGet use trusted publishing (OIDC), never a committed or printed key. This guard is active in every Nihil mode.
- Never run a catastrophic, unrecoverable command.
rm -rf /or~,mkfs,dd of=/dev/....git reset --hard, force-push, andgit clean -fare allowed — they are recoverable from the reflog or the remote.
If you are blocked, it is one of these two. Everything else flows.
How to raze
- Scope. Restate what you were asked to transform (
$ARGUMENTS— a path, a subsystem, or the whole repo). If unscoped, ask once, then proceed. - Inspect, briefly. Learn the build/test commands, the public surface, and how validation runs — enough to verify your own work, not enough to stall.
- Pick the smallest coherent transformation from the change-magnitude ladder in
skills/nihil/SKILL.mdthat reaches a coherent end state. Smallest coherent — not smallest timid. A no-op is a valid verdict if the code already earns its existence. - Transform, write-capable, end to end. Make the change. Do not leave a degraded intermediate state. Do not stop because the change is large, the architecture is familiar, public APIs break, or deletion feels aggressive. Stop only for: a secret or catastrophic brake, missing credentials/infrastructure, an unresolved external product decision, genuinely unsafe uncertainty, or explicit interruption.
- Verify with the real artifact. Build, run the tests, run the thing. Compare against a baseline you captured before editing — never report a failure you caused as preexisting.
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.
- 16d ago First seen · 92 lines · 103 tokens per session scan C f2d1ea68675e
raze is a skill published in the GitHub repository ANcpLua/ancplua-claude-plugins (6 stars, last pushed 5d ago), licensed MIT. It adds 103 tokens to every session and 1,080 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-01.
Other skills, from other repositories
create-pr
Creates a GitHub PR with a Linear-ticket-prefixed title and a decision-led, narrative description for Prisma 8. Use when the user wants to create a pull request, open a PR, or submit changes for review.
git-commit
Generate well-formatted git commit messages following conventional commit standards.
repo-sync
Manage automatic Git syncing on macOS with repo-sync. Use only when the user explicitly invokes /repo-sync.
mcore-split-pr
Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.
contributing
How to contribute to evlog, covering commit and PR conventions, changesets, the Definition of Done, testing rules, and the authored skills that walk through building a new adapter, enricher, framework integration, or map rule. Load this for any question about contributing, opening a PR, or adding something to the…
git-commit-writer
Trigger when the user asks to write a commit message, generate a commit from staged changes, review a diff before committing, or says "/commit". Reads git diff --staged and produces an opinionated, convention-matching commit message.