The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/gather.py executes git and gh (GitHub CLI) commands using subprocess.run to fetch repository metadata, commit logs, and pull request activity.
[DATA_EXPOSURE]: The skill accesses sensitive local directories and files to retrieve user interaction history, specifically reading from ~/.claude/projects, ~/.codex/sessions, and the SQLite database at ~/.local/share/opencode/opencode.db.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external sources (GitHub PR titles/messages) and local history which could contain embedded instructions intended to manipulate the agent's output.
Ingestion points: Data is gathered from GitHub API results, git command outputs, and local agent session logs via scripts/gather.py.
Boundary markers: No technical delimiters are used to separate gathered data from the agent's instructions, though the skill body provides formatting guidelines.
Capability inventory: The skill has the ability to execute shell commands (git, gh) and read files from the user's home directory.
Sanitization: The gathered text content is processed without explicit sanitization or filtering for potential instruction-like patterns.

standup