researching-a-topic
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes the
webfetchtool to retrieve and process content from external URLs and thegh_greptool to search through GitHub repositories.- [COMMAND_EXECUTION]: Thebashtool is made available to the agent for data processing tasks, which presents a potential risk if the tool is used on maliciously crafted data retrieved from the web.- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes untrusted data from external sources without sufficient safety boundaries. - Ingestion points: External content enters the agent context through
webfetchandgh_grep(SKILL.md). - Boundary markers: The instructions lack explicit requirements for the agent to use delimiters or 'ignore embedded instructions' markers when processing retrieved text.
- Capability inventory: The skill allows for command execution via
bashand file modification via thewritetool. - Sanitization: No sanitization, escaping, or validation steps are defined for the data retrieved from external sources before it is analyzed or used.
Audit Metadata