search-history
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use shell commands (
grep,find) to search through local chat history files stored in JSONL format. - [DATA_EXPOSURE]: The skill accesses sensitive local directories (
~/.claude/and~/.codex/) to read conversation logs. This is the primary intended function of the skill and is handled as read-only access. - [INDIRECT_PROMPT_INJECTION]: As the skill processes historical chat data, it could potentially encounter past malicious instructions (e.g., in logs of previous security testing). The skill includes a
disable-model-invocation: trueconfiguration and instructions to only output specific context fragments, which helps mitigate the risk of the agent accidentally executing instructions found within the logs.
Audit Metadata