skills/harshanandak/forge/smith/Gen Agent Trust Hub

smith

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests untrusted data from an external issue tracker and possesses command execution capabilities. Maliciously crafted issue descriptions could attempt to influence the agent's behavior during the orchestrated triage-to-ship workflow.\n
  • Ingestion points: The skill retrieves issue content and metadata using forge issue ready --json in SKILL.md.\n
  • Boundary markers: There are no explicit delimiters or instructions provided to isolate ingested issue text from the agent's core instructions.\n
  • Capability inventory: The skill uses Bash (restricted to the forge CLI) and manages several sub-skills (plan, dev, validate, ship, review, verify) to perform automated development tasks.\n
  • Sanitization: No specific sanitization or validation of the ingested issue data is mentioned in the orchestration logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 10:07 AM
Security Audit — agent-trust-hub — smith