forge-infrastructure

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Node.js script located at ../../../.fullstack-forge/runtime/cli/src/composition-entry.js using the system's node interpreter to perform 'runtime composition'.
  • [PROMPT_INJECTION]: This skill functions as an 'adapter' that redirects the agent to load instructions from a canonical playbook file (../../../.fullstack-forge/skills/forge-infrastructure/SKILL.md) and additional paths specified in a local configuration file (.forge/composition.json). This pattern creates an indirect prompt injection surface where the agent's behavior is determined by data in the local environment.
  • Ingestion points: .forge/composition.json (local file), ../../../.fullstack-forge/skills/forge-infrastructure/SKILL.md (external file relative to skill directory).
  • Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore potential commands within the loaded composition data.
  • Capability inventory: File system read access and shell command execution via node.
  • Sanitization: No evidence of sanitization or validation of the paths or content provided by the composition JSON or canonical playbook.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 06:48 PM
Security Audit — agent-trust-hub — forge-infrastructure