session-cleaner
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFE
Full Analysis
- COMMAND_EXECUTION (SAFE): The skill utilizes
pkillandopenclawCLI commands. These are used strictly to manage the lifecycle of its own gateway service and sessions. This is standard behavior for a system management utility. - DATA_EXFILTRATION (SAFE): The Python script modifies session files located in
~/.openclaw/. Analysis confirms it only performs local file operations (read, write, rename) and contains no networking code or exfiltration logic. - REMOTE_CODE_EXECUTION (SAFE): No remote dependencies or external script downloads were found. All execution is limited to the provided local Python script and standard system commands.
- PROMPT_INJECTION (SAFE): The instructions in
SKILL.mdare purely functional and do not contain attempts to override agent behavior, extract system prompts, or bypass safety filters.
Audit Metadata