verification-gate
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to run development commands and parse raw terminal output, logs, and code diffs, introducing a potential surface for indirect prompt injection if an adversary injects malicious commands or instructions into test logs, file diffs, or source comments.
- Ingestion points: Raw terminal stdout/stderr logs from testing frameworks (
npm run build,npm test,playwright) and list/content outputs fromgit diffas described inSKILL.md. - Boundary markers: Absent; no special delimiters or instructions are specified to separate the external command output text from the agent's internal instruction context.
- Capability inventory: High capability to run local shell scripts, toolchains (
npm,npx,cargo check), search utilities (grep), and version control systems (git) across all phases. - Sanitization: Absent; the agent reads raw or truncated (
tail) command outputs directly to populate the final verification report scorecard without pre-filtering.
Audit Metadata