shell-history
Warn
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of shell commands, specifically 'atuin' and 'grep', to retrieve and filter terminal history data.- [DATA_EXFILTRATION]: The skill accesses shell command history, which is a highly sensitive data source often containing secrets such as passwords, tokens, and environment variables. The instructions specifically direct the agent to write this history into temporary files in '/tmp', creating a potential risk of local data exposure if these files are not properly secured or deleted.- [PROMPT_INJECTION]: The skill ingests untrusted terminal history logs which could contain malicious instructions from previous commands or their outputs, leading to indirect prompt injection.
- Ingestion points: Command history output from 'atuin search' in SKILL.md.
- Boundary markers: Not present; the instructions do not define clear delimiters to separate data from instructions.
- Capability inventory: The skill has access to shell execution via bash.
- Sanitization: None identified; the history content is grepped and processed directly without filtering for potential prompt injection patterns.
Audit Metadata