enterprise-search-strategy
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and synthesize data from various internal sources that are potentially attacker-influenced.
- Ingestion points: Sources include chat archives (Slack, Discord, Teams), issue trackers (Jira, Linear), and code repository comments/PRs, as defined in Step 1 of the procedure.
- Boundary markers: The instructions lack specific guidance for the agent to use delimiters or 'ignore instruction' markers when processing content from these sources.
- Capability inventory: The strategy involves reading files and capturing hits, which interacts with the agent's file system and search tools.
- Sanitization: There are no instructions for sanitizing or escaping the content retrieved from these internal indexes before it is synthesized into an evidence brief.
Audit Metadata