session-search
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a bash script (
session-search.sh) and markdown snippets to execute system commands such asjq,rg,grep, anddu. These commands use arguments derived from user input, specificallyjob_idand searchpattern. - [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection (Category 8) because it processes untrusted data from past session logs.
- Ingestion points: Conversation logs and job prompts located in
/job/logs/(specifically.jsonlandjob.mdfiles). - Boundary markers: None; there are no instructions to the agent to disregard instructions within the logs.
- Capability inventory: The skill provides the ability to read and search local log files using shell utilities.
- Sanitization: None; log content is presented to the agent without filtering for potentially malicious commands or instructions.
Audit Metadata