Skip to content
Back to skills

Audit

ASecurity

Security & dependency audit using Gemini's grounded Google Search.

  • 3 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added May 26, 2026
developmentgobashnodesecurity

Works with

  • cli

Security analysis

A100/100

Scanned May 27, 2026

npx -y skills add KingGyuSuh/gemini-search-cc --skill audit --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Audit?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Audit
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/kinggyusuh-audit/badge)](https://www.skillsdirectory.com/skills/kinggyusuh-audit)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
description: Security & dependency audit using Gemini's grounded Google Search.
argument-hint: '<package or technology>'
disable-model-invocation: true
allowed-tools: Bash AskUserQuestion Glob Read
---
Perform a security and dependency audit using Gemini CLI's grounded Google Search.

Use this to check packages, libraries, or technologies for vulnerabilities,
deprecations, and compatibility issues before adopting or upgrading them.

Raw slash-command arguments:
`$ARGUMENTS`

## Steps

1. If `$ARGUMENTS` is empty, use `AskUserQuestion` to ask what to audit.

2. **Gather context** (optional but recommended):
   - Use `Glob` and `Read` to check for `package.json`, `requirements.txt`, `Cargo.toml`,
     `go.mod`, or similar dependency manifests in the workspace.
   - If found, extract the relevant package version to enrich the audit query.

3. Run the companion script:
   ```bash
   node "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-companion.mjs" audit "$ARGUMENTS"
   ```

4. If the command fails:
   - Run `node "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-companion.mjs" setup` to diagnose.
   - Show the diagnosis to the user.

5. Return the output **verbatim**.

6. After the results, provide a **Audit Verdict**:
   - **Risk Level:** Critical / High / Medium / Low / Clear
   - **Action Required:** Yes / No
   - **Summary:** One-sentence recommendation (proceed, update, replace, or avoid)

## Constraints
- Do not paraphrase the Gemini output.
- Always err on the side of caution — if uncertain, recommend the user investigate further.
- Do not auto-fix or auto-update any dependencies. Only advise.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.