The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run a bundled Python script (fix_surrogates.py) to perform maintenance on chat history files. The script performs local file system operations, including reading, writing, and creating backups (.bak) of files within the ~/.claude/projects/ directory. These actions are necessary for the skill's stated purpose of repairing encoding issues.
[SAFE]: The logic in the Python script is transparent and follows best practices for handling encoding errors (using surrogateescape and surrogatepass). It does not use any dangerous functions like eval() or exec(), nor does it perform any network operations.
[DATA_EXFILTRATION]: No network calls or data exfiltration patterns were detected. All processing is done locally on the user's machine.
[PROMPT_INJECTION]: The instructions in the skill do not contain any patterns intended to bypass safety filters or override the agent's core behavioral constraints.

fix-jsonl-surrogates