review-integration
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze untrusted external data (integration packages and PR diffs), which presents a risk of indirect prompt injection if those files contain malicious instructions.
- Ingestion points: The skill reads a wide range of files from the package under review, including
manifest.yml,cel.yml.hbs, ingest pipelines, field definitions, and documentation (as defined in SKILL.md Step 2). - Boundary markers: The instructions lack explicit requirements for using boundary markers or 'ignore' instructions when interpolating the content of these untrusted files into the agent's context.
- Capability inventory: The agent can execute shell commands using the
elastic-packageCLI (SKILL.md Step 5) and write review results to the local filesystem attmp/integration-review.md. - Sanitization: There is no specified mechanism for sanitizing or escaping the content of the ingested files before processing, which could allow embedded instructions to influence the agent's behavior.
Audit Metadata