The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes Bash to execute standard Git and GitHub CLI commands, along with embedded Python scripts for parsing JSON and formatting strings.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata from an external environment variable and interpolates it directly into repository artifacts (commits and PR descriptions).
Ingestion points: Reads the $PYLOT_REQUESTORS environment variable in SKILL.md.
Boundary markers: None present to distinguish attribution content from instructions.
Capability inventory: Invokes git commit to modify history and gh pr edit to modify pull request content.
Sanitization: The skill performs minimal sanitization by stripping angle brackets (<, >) from names, but does not validate the content against malicious patterns or prompt injection payloads.

requestor-attribution