Metadata Mismatch Skill
A skill with mismatched metadata declarations.
Slack Notifications
Send structured status updates to a Slack channel via Incoming Webhooks. Say "notify the team in Slack", "send a Slack message to #channel", or "post update to #general" to trigger this skill.
Legitimate Redirection Skill
A skill that uses output redirection in a legitimate context (binary existence check).
Secure File Summary
Summarize local files and write output to a designated directory. No network access. Say "summarize the files in [path]", "give me a summary of [file]", or "create a report from [folder]" to trigger this skill.
No Network But Curls
A skill that declares no network egress but makes curl calls.
Injection Only Skill
A skill that contains prompt injection patterns but no other issues.
Undeclared Env Skill
A skill that uses environment variables without declaring them.
Claude Code Security
Scan a local codebase or GitHub repository for security vulnerabilities using Claude Code Security. Returns severity-scored findings and patch suggestions for human review. Say "scan [path] for vulnerabilities", "run a security review on [path]", or "find vulnerabilities in [path]" to trigger this skill.
What Would Elon Do
Get Elon's perspective on any topic. Ask me anything and I'll channel his mindset.
Clean Example
A properly declared skill for testing. Fetches data from the declared API endpoint.
Audit Logger
Automatically log all skill invocations to a local append-only audit file. This skill hooks into the session lifecycle and is not invoked directly by the user.