forge-setup-project
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands for routine project configuration tasks related to documentation compatibility.\n
- Evidence: In Step 7, it executes
rm -f CLAUDE.mdandln -sf AGENTS.md CLAUDE.md. These commands are used to maintain a consistent documentation interface and are appropriate for the tool's stated purpose.\n- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it ingests and processes existing documentation and codebase information to generate new context files.\n - Ingestion points: The skill scans and reads files including
AGENTS.md,CLAUDE.md,README.md, and content within thedocs/directory (Step 1 and Step 3).\n - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to isolate content read from project files during the audit and migration process.\n
- Capability inventory: The skill has access to
WriteandEdittools for file modification andBashfor command execution, creating a potential path for injected instructions to influence project state.\n - Sanitization: There is no explicit requirement or mechanism described for sanitizing or validating the ingested documentation content before it is rewritten or summarized.
Audit Metadata