Pliperkiller

Audit Figure — Security skill for Claude Code

Security community

Audit a figure, or the code that generates it, against the scientific-figures rules.

How to install Audit Figure

Installs to ~/.claude/skills/pliperkiller-research-agent-audit-figure/SKILL.md

Terminal

mkdir -p ~/.claude/skills/pliperkiller-research-agent-audit-figure && curl -fsSL https://raw.githubusercontent.com/Pliperkiller/research-agent/HEAD/commands/audit-figure.md -o ~/.claude/skills/pliperkiller-research-agent-audit-figure/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What Audit Figure does

description: Audit a figure, or the code that generates it, against the scientific-figures rules argument-hint: [path to a .py or .ipynb, or a description of the figure]

Audit the figure given in `$ARGUMENTS` using the `scientific-figures` skill.

If $ARGUMENTS is a path to a .py or .ipynb, run python scripts/figure_lint.py from the skill.
Read references/checklist.md and walk the items the linter cannot decide on its own (R1, R6, R7, R10, cross-panel consis

Alternatives in Security

Skill Coverage Audit — Trace codepaths in diffs, map against tests, auto-generate missing coverage — use before shipping PRs 2.8k ★
Unifly Audit — Security audit of UniFi configuration 248 ★
Commit Hook Checklist — Audit commit-hook automation as gates against AI slop and broken commits (Node.js, Go, polyglot) 168 ★