forge-deployment

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Node.js script using node ../../../.fullstack-forge/runtime/cli/src/composition-entry.js. This is part of the core functionality to resolve the runtime composition of the deployment skill.
  • [DATA_ACCESS]: The skill accesses local files including ../../../.fullstack-forge/skills/forge-deployment/SKILL.md and .forge/composition.json. This access is required to retrieve the canonical instructions and repository-specific configuration.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it dynamically loads instruction paths defined in the repository's .forge/composition.json file.
  • Ingestion points: .forge/composition.json (local repository file).
  • Boundary markers: Not explicitly defined in the adapter.
  • Capability inventory: Execution of local node scripts and reading filesystem paths.
  • Sanitization: The skill includes a 'forge doctor' and 'forge update' mechanism for integrity checks, though it relies on the canonical framework for validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 06:48 PM
Security Audit — agent-trust-hub — forge-deployment