The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes build and test commands defined in the project configuration files (CLAUDE.md or AGENTS.md) to verify changes before shipping.\n- [COMMAND_EXECUTION]: Uses the grep utility through lib/quote-gate.md to validate that quotes and citations in learning artifacts match the source code.\n- [EXTERNAL_DOWNLOADS]: Synchronizes the local environment with remote changes using git pull during the pre-flight phase.\n- [DATA_EXFILTRATION]: Transfers code and metadata to remote repositories using git push and the GitHub CLI (gh pr create), which are standard operations for code delivery to the project's configured remote hosts.\n- [PROMPT_INJECTION]: The skill processes untrusted project data (briefs, configuration files) which could theoretically influence agent behavior, though this is managed as part of the intended project context.\n
Ingestion points: Reads test commands and scope definitions from CLAUDE.md, AGENTS.md, and docs/briefs/**.\n
Boundary markers: None explicitly defined for the project-level command execution.\n
Capability inventory: Includes full repository read/write access, shell execution, and network access via Git/GitHub.\n
Sanitization: No specific sanitization is performed on project-defined build/test strings.

ship