forge
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes various forms of untrusted external data, creating an attack surface for indirect prompt injection where malicious instructions could be embedded in data sources.
- Ingestion points: Repository text, fetched web pages, issue content, and tool output are ingested into the agent context.
- Boundary markers: The instructions do not define specific delimiters or boundary markers to isolate external data from the primary prompt instructions.
- Capability inventory: The skill executes CLI commands (forge), manages local state files in
.forge/, and routes to other functional engines. - Sanitization: The instructions explicitly mandate treating all external sources as untrusted and require revalidation of roots, revisions, hashes, and artifacts during every resume or verification step.
Audit Metadata