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 OutlineDriven/odin-claude-plugin --skill guillotinegit clone --depth 1 https://github.com/OutlineDriven/odin-claude-pluginWrote 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/outlinedriven/odin-claude-plugin/guillotine)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/guillotine"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/guillotine/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/outlinedriven/odin-claude-plugin/guillotine"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/guillotine.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 | $0.00045 | $0.01244 |
| Opus 5 | $0.00023 | $0.00622 |
| Sonnet 5 | $0.00009 | $0.00249 |
| Haiku 4.5 | $0.00005 | $0.00124 |
Grade A, and why
guillotine 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 2d 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guillotine
Remove weight the product does not use. Interior cleanup needs proof, not sentiment.
Contract
| Field | Bound contract |
|---|---|
| Trigger | A repository or subsystem needs dead, duplicate, superseded, or generated residue removed across more than one artifact class. |
| Authority | Reversible local: writes only VCS-tracked interior artifacts (edits and deletion); rollback is version control. No remote mutation. No history rewrite, data migration, credential change, or deletion of untracked or critical data. |
| Side effect | VCS-tracked artifacts are deleted or consolidated; live consumers are migrated to the surviving owner; generated surfaces are regenerated. The observable boundary is unchanged. |
| Done | Every removed artifact has a proof of non-necessity, no live route to the old path remains, the observable boundary passes, and the repository gate is green. |
Inputs
- The subsystem or repository region to purge. Named explicitly or derived from the observed debt surface.
- The observable boundary: the commands, public routes, generated outputs, or user scenarios that must remain unchanged. Required; if no boundary can be named, the skill stops before editing.
- The repository-native gate command that defines "passes". Required, or discoverable in the repo.
Procedure
- Name the subsystem and its observable boundary. List the commands, public routes, generated outputs, or user scenarios that must remain unchanged. Done when: the subsystem is named and every boundary element is listed, or the skill stops because no provable boundary exists.
- Build a candidate ledger. For each code path, document, test, dependency, configuration entry, workflow, or generated file, record its claimed job, live callers or readers, source of truth, and evidence that it is dead, duplicate, superseded, or reproducible. Done when: every candidate has an entry with its job, live consumers, source of truth, and evidence category named.
- Classify every candidate into one of five categories. Dead: no live caller, route, import, loader, build step, or documented consumer. Duplicate: another artifact owns the same fact or operation and all consumers can move to it. Superseded: the active path covers the required contract and no supported path selects the old one. Generated residue: the repository generator reproduces the required output and does not select this copy. Keep: evidence still names a live job, even when the artifact looks ugly. Done when: every candidate carries exactly one classification and the five categories are exhaustive with no candidate unclassified.
- Prove the cut from both directions. Search from each candidate to its consumers, then from each required consumer back to its owner. Run a focused behavioral, mutation, build, or generation probe when static references cannot prove necessity. Done when: every candidate has a bidirectional proof: consumer-to-candidate search and candidate-to-consumer search both confirm the classification, or a probe was run that settled the ambiguous case.
- Stop for approval if the proposed cut reaches an observable API, user command, stored data, history, remote state, untracked file, credential, or critical target. Show the exact consequence and a recoverable alternative. Interior VCS-tracked debt with a proven unchanged boundary does not need a second approval. Done when: every approval-gated target is identified with its consequence and alternative stated, or no target requires approval and the cut proceeds.
- Cut one dependency-closed slice at a time. Move live consumers to the single surviving owner, then delete the obsolete implementation, tests that protect only obsolete behavior, dependency entries, configuration, workflow steps, and generated copies. Do not leave aliases, shims, dual paths, tombstones, or "deprecated" wrappers. Done when: the slice is cut with all consumers migrated to one owner and no alias, shim, dual path, tombstone, or deprecated wrapper left behind.
- Regenerate every repository-owned surface affected by the cut. Reject a generator that recreates the removed residue; fix its source selection instead of deleting its output again. Done when: every affected generated surface is regenerated and the generator does not reproduce the removed artifact.
- Exercise the observable boundary and run the repository-native gate. If behavior changes, restore the smallest failed slice from VCS, correct its ledger classification, and continue with the remaining proven cuts. Done when: every named boundary element passes and the repository gate exits green, or the failed slice is restored and reclassified.
- Re-run route, import, registration, and source-of-truth searches. Done when: the old path is absent from every search and every required consumer resolves to one live owner.
What ships with it
1 file 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.
- 2d ago Changed · -33 tokens per session 53932b178796
- 4d ago First seen · 52 lines · 78 tokens per session scan A 4a5042dbff7e
guillotine is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (35 stars, last pushed yesterday), licensed Apache-2.0. It adds 45 tokens to every session and 1,244 once invoked, about $0.0002 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 2026-09-04.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…