Exorcise — Security skill for Claude Code
Audit the current changeset (or a PR) against its stated intent, then apply — revert hunks nothing in the intent reaches, inline single-caller abstractions, swap new code for existing helpers, move po.
How to install Exorcise
Installs to ~/.claude/skills/jacquardlabs-exorcist-exorcise/SKILL.md
mkdir -p ~/.claude/skills/jacquardlabs-exorcist-exorcise && curl -fsSL https://raw.githubusercontent.com/jacquardlabs/exorcist/HEAD/commands/exorcise.md -o ~/.claude/skills/jacquardlabs-exorcist-exorcise/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Exorcise does
description: Audit the current changeset (or a PR) against its stated intent, then apply — revert hunks nothing in the intent reaches, inline single-caller abstractions, swap new code for existing helpers, move point-of-use fixes to the entry point, delete what the change made unnecessary. Pass the intent as text, a PR number or URL, or nothing to read it from the branch. Pass a séance register path to work a register instead. allowed-tools: Bash, Read, Edit, Write, Glob, Grep, Task, AskUser
Alternatives in Security
- Skill Code Review — Expert multi-AI code review with inline PR comments — use for thorough quality and security analysis 2.8k ★
- Maestro Knowledge — Intent-driven knowledge-store and Run knowledge lifecycle management — audit/prune, stage candidates (with sig 530 ★
- Bountyforge — All-round bug bounty skill for Claude Code parallelized agents for smart contract audits (EVM, Move, Solana, T 399 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Gaia Audit
Audit memory, wiki, and auto-loaded files for duplication, conflicting instructions, and stale content. The de
Context Gardening
Audit CLAUDE.md, its linked docs, and .claude/memory/ against the patterns from "A good AGENTS.md is a model u
Review Implementation
Independent implementation audit of uncommitted code by Codex CLI against a plan doc, then apply agreed fixes
App Audit
Grade an existing app against the Mobify Studio House KB and Axiom standards — fan out the specialist auditors
Ship Check
End-of-session production audit with auto-merge. Reconstructs session intent, then RUNS the full review chain
UI Refine
Iterative UI refinement: audit existing UI, select issues to fix, apply fixes, re-audit. Provide file paths or
Related Agents
Intent Tracer
Traces every hunk of a diff to the stated intent. Returns a JSON array of findings — hunks nothing in the inte
Light Review Step3
Stage 3 of the Delphi review pipeline — verify code fixes hold up, revert the ones that don't, then compile. F
Cleanup Phase 3 Survey
Diagnoses the folder structure in phase 3 of a codebase cleanup and produces the move plan for the user to app