The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/release.py script executes external command-line tools acli (Jira) and gog (Google Sheets) using Python's subprocess module. These calls utilize structured argument lists rather than raw shell strings, which is a best practice that prevents shell injection vulnerabilities.
[EXTERNAL_DOWNLOADS]: The skill documentation in references/config.md provides instructions for installing the gog CLI tool from a third-party source (gogcli.sh) via Homebrew. It also fetches release data and team mappings through official Jira and Google Sheets APIs.
[PROMPT_INJECTION]: The skill ingests data from external sources like Jira issue descriptions and Google Sheets cells to generate release updates and Slack announcements. It includes built-in mitigations to manage this surface.
Ingestion points: Data is retrieved from Jira work items and Google Sheet ranges via CLI tools.
Boundary markers: The SKILL.md principles mandate wrapping Slack messages in triple-backtick code blocks (```slack) to maintain clear context separation.
Capability inventory: The skill has the capability to execute shell commands (acli, gog) and read local configuration files.
Sanitization: Output is processed through Python scripts that use json.dumps for structured data and enforce strict templates for Slack announcements.

rhdh-release