skill-search

Warn

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill performs git clone operations and web fetching from multiple third-party GitHub repositories and external websites (e.g., skillsmp.com) that are not on the trusted vendors list.
  • [REMOTE_CODE_EXECUTION]: The skill automates the installation of arbitrary code by copying files from remote repositories directly into the agent's persistent skill directories (~/.claude/skills/ and ~/clawd/skills/). This allows for the execution of third-party code that has not been verified or sanitized.
  • [COMMAND_EXECUTION]: The skill utilizes shell commands including git, find, grep, and cp -r to manage and install content from external sources.
  • [PROMPT_INJECTION]: 1. Ingestion points: Downloads SKILL.md files from various community repositories and fetches content from external sites. 2. Boundary markers: No delimiters or ignore instructions directives are used when processing this remote content. 3. Capability inventory: The skill has significant capabilities including file system modification, command execution, and network access. 4. Sanitization: There is no evidence of validation or sanitization of the downloaded code or text before installation or presentation to the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 13, 2026, 07:00 AM