app-and-add-on-lifecycle-advisor
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses the
webtool to retrieve documentation from official Splunk domains, includinghelp.splunk.com,dev.splunk.com, andsplunkdev.page. These are authoritative vendor resources used for providing cited lifecycle guidance. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as release notes, AppInspect reports, and vendor documentation. To mitigate injection risks, the instructions explicitly command the agent to treat all retrieved and supplied content as untrusted evidence and not as executable instructions. The skill's capabilities are restricted to advisory, non-mutating actions.
- Ingestion points: Release notes, app/vendor documentation, Splunkbase records, and AppInspect results (SKILL.md).
- Boundary markers: Present; the skill instructs the agent to treat retrieved content as untrusted evidence and separate documented guidance from environment observations (SKILL.md).
- Capability inventory: Access to the
webtool for documentation lookup; no file-write, configuration change, or code execution capabilities (SKILL.md). - Sanitization: Present; instructions specify using sanitized release notes and evidence-bounded advice (SKILL.md).
Audit Metadata