Skills
skills/alcyone-labs/agent-skills/skill-forge/Gen Agent Trust Hub

skill-forge

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references an installation script hosted on the author's official GitHub repository (Alcyone-Labs/agent-skills).
  • [REMOTE_CODE_EXECUTION]: The Gemini adapter provides instructions for the user to download and execute a remote script via curl | bash for updates.
  • [COMMAND_EXECUTION]: The skill uses npx to execute logic from the @alcyone-labs/agent-skills package and manages files within the .agents/ and ./skills/ directories.
  • [PROMPT_INJECTION]: The skill ingests user input via arguments to generate skill structures, which creates a surface for indirect prompt injection.
  • Ingestion points: Arguments passed to the droid and opencode adapters.
  • Boundary markers: None present to distinguish user input from instructions.
  • Capability inventory: File writing and command execution via the agent-skills CLI.
  • Sanitization: No visible input sanitization or validation before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 04:15 AM