The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell scripts and CLI tools to manage the repository backlog.\n
The skill uses git, grep, awk, and sort within scripts located in the scripts/ directory to query and update the TODO.md file.\n
The frontmatter allowed-tools configuration Bash(sh ${CLAUDE_SKILL_DIR}/scripts/*) limits the agent's ability to execute arbitrary commands, restricting it to the provided scripts.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from repository files which could be manipulated to influence agent behavior.\n
Ingestion points: The skill reads content from TODO.md and STORIES.md at the repository root using grep, awk, and git show (as seen in SKILL.md, scripts/list.sh, and scripts/stories.sh).\n
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when presenting the file content to the agent.\n
Capability inventory: The skill has the ability to read and write repository files, execute local shell scripts, and perform various git operations (log, show, blame, etc.).\n
Sanitization: No sanitization or validation of the content within the TODO.md or STORIES.md files is performed before the content is ingested into the agent context.

todo