how
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes arbitrary content from the codebase being analyzed.
- Ingestion points: Codebase files are ingested into the agent context via
glob,grep, andviewtools as described inSKILL.mdandexplorer-prompt.md. - Boundary markers: The subagent templates in the
references/directory use standard markdown headers to organize input but lack explicit delimiters or instructions to ignore potential commands embedded within the code being read. - Capability inventory: The skill possesses significant capabilities including the
bashshell and thetasktool for spawning subagents, which could be leveraged if an attacker successfully injects instructions into the codebase. - Sanitization: The skill does not appear to perform any sanitization or filtering of the content retrieved from files before synthesizing findings or presenting them to the user.
Audit Metadata