duck-triage
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates on potentially untrusted external inputs such as bug reports, logs, and reproduction artifacts, which is an inherent characteristic of triage tools.
- Ingestion points: The workflow in
SKILL.mdrequires collecting "repro artifacts", "logs", and "bug reports" to perform severity classification. - Boundary markers: The skill does not define specific delimiters (like XML tags or triple backticks) to separate untrusted data from instructions, though it uses structured output prefixes like
missing evidence:. - Capability inventory: The skill is strictly instructional and does not invoke tools for file system modification, network operations, or shell execution.
- Sanitization: No explicit sanitization or filtering of input data is defined, but the skill includes a "Safety carve-outs" policy in
references/GUARDRAILS.mdwhich commands the agent to refuse any changes that weaken security controls or trust-boundary validation.
Audit Metadata