The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell commands using gh (GitHub CLI) and git to retrieve repository metadata, issue lists, pull request statuses, notifications, and commit history. These operations are essential for its primary function of generating standup reports.
[DATA_EXFILTRATION]: The skill reads repository activity data and local commit logs. While this data is used to generate a local report, the skill instructions mention integration with external communication tools like Slack or Teams via formatting templates, which involves the user or agent transferring this summarized data to external platforms.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves untrusted data from external sources (GitHub issues, PR comments, and git commit messages) and processes this content using sub-agents for classification and impact analysis. A malicious actor could craft issue titles or commit messages designed to influence the agent's report or behavior.
Ingestion points: gh issue list, gh pr list, gh api notifications, and git log output.
Boundary markers: Absent; the sub-agent prompts interpolate the retrieved data without explicit delimiters or "ignore instructions" warnings.
Capability inventory: Shell execution for data gathering (gh, git), date calculation, and summarized report generation.
Sanitization: None; the skill directly processes text retrieved from GitHub and the git history.

gh-daily