repo-skill-installer

Warn

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's primary workflow involves fetching content from remote sources, specifically targeting GitHub repositories and skills.sh URLs to install them locally.
  • [REMOTE_CODE_EXECUTION]: By downloading and placing executable skill components into the .agents/skills directory, the skill facilitates the execution of remote code. The instructions specifically mention resolving and copying scripts from external origins, which can lead to the execution of untrusted logic.
  • [PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by processing external, unverified data that may contain malicious instructions.
  • Ingestion points: External repositories and skills.sh URLs are ingested into the local repository context.
  • Boundary markers: No explicit markers or delimiters are used to isolate or ignore instructions contained within the downloaded content.
  • Capability inventory: The skill possesses the capability to write files to the local directory .agents/skills/ and modify repository configuration files including skills.sources.json and skills-lock.json.
  • Sanitization: There is no evidence of sanitization or integrity verification for the remote content before it is integrated into the system.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 14, 2026, 03:31 PM
Security Audit — agent-trust-hub — repo-skill-installer