The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted content from documentation files (README.md, ARCHITECTURE.md, CONTRIBUTING.md, CLAUDE.md, and project-specific .md files) to perform factual audits and voice polishing. * Ingestion points: Documentation files discovered via the find command in Step 1 and read in Step 2. * Boundary markers: Absent. The skill does not use specific delimiters or instructions to ignore potential commands embedded within the documentation files being processed. * Capability inventory: The agent has the ability to edit files, commit changes to the repository, push to remote branches, and update GitHub Pull Request descriptions. * Sanitization: Absent. There is no evidence of validation or sanitization of the content read from external documentation files before it is processed by the AI.
[COMMAND_EXECUTION]: The skill executes multiple local binaries for lifecycle management and telemetry. * Evidence: Calls to gstack-update-check, gstack-config, and gstack-telemetry-log located in ~/.claude/skills/gstack/bin/. * Evidence: Use of the open command to launch a browser to an external URL, though this is gated by a user confirmation step.

document-release