The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands like cp, Copy-Item, and git mv to set up the project structure and install its local utility scripts. These actions are limited to the user's project directory and the skill's own installation path.
[DATA_EXFILTRATION]: The skill audits local repository files to generate documentation reports. It operates entirely locally and lacks network access capabilities for transmitting data externally.
[PROMPT_INJECTION]: The migration workflow scans existing project documentation, creating an indirect prompt injection surface. Ingestion points: SKILL.md (scanning all repository .md files). Capability inventory: local file movement and document structure organization. Boundary markers and sanitization are absent, but the risk is managed by the requirement for user review of all proposed migration steps.
[REMOTE_CODE_EXECUTION]: Included Python tools (search-yaml.py, validate-yaml.py) perform local parsing and validation using standard libraries; no patterns of remote code fetching or execution were identified.

cs-onboard