gadd-verify
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect prompt injection vulnerability surface detected in the data processing workflow.
- Ingestion points: The skill ingests untrusted text from Work Item bodies, ledger files, GitHub Pull Request metadata, and external issue trackers.
- Boundary markers: Absent. The instructions do not define delimiters or headers to isolate untrusted data from the agent's instructions.
- Capability inventory: The skill performs file system reads and writes to local project files (ledger.yml, verification.md) and makes network calls to read pull request metadata. It does not execute arbitrary shell commands.
- Sanitization: The instructions do not require the sanitization or validation of content retrieved from external trackers or pull request descriptions.
Audit Metadata