The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for indirect prompt injection as it processes data from external repositories and project files.\n
Ingestion points: The skill executes gh api user/orgs and gh repo list to retrieve repository metadata, and uses rg and fd to scan project source files for internationalization strings.\n
Boundary markers: No explicit delimiters or guardrails are defined to ensure the agent ignores malicious instructions potentially embedded in repository names, descriptions, or source code contents.\n
Capability inventory: The skill has significant capabilities, including the ability to perform git commit and git push operations and modify local source files.\n
Sanitization: The protocol does not specify sanitization or validation steps for the data retrieved from GitHub APIs or local files before processing.\n- [SAFE]: The skill follows security best practices by requiring that sensitive internal documentation, such as legal risk assessments, be stored in a .gitignore protected directory (_after-care/) to prevent accidental public disclosure. All shell commands are consistent with standard maintenance procedures.

full-after-care