The Agent Skills Directory

[SAFE]: The skill demonstrates security-conscious design by providing a 'gitx scan' utility that identifies exposed credentials in the working tree and history while redacting their values and warning the user before including sensitive files (e.g., .env, .pem, .key) in commits.\n- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by reading untrusted data from GitHub issues and comments (Ingestion points: 'gh issue view' in SKILL.md). Boundary markers are present in the form of explicit instructions to treat external issue content as untrusted reference material and to disregard any embedded instructions that conflict with repository safety rules. Capability inventory: the skill is capable of modifying the repository, using the GitHub CLI, and executing local test runners (SKILL.md). Sanitization: the secret scanning workflow redacts identified credentials from output.\n- [COMMAND_EXECUTION]: The skill invokes standard local development commands and test runners (e.g., 'npm test', 'pytest', 'cargo test', 'make test') to validate changes (SKILL.md). These operations are restricted to existing project infrastructure and occur within the developer's local environment.

gitx