The Agent Skills Directory

[DYNAMIC_EXECUTION]: The read-config.sh utility executes an embedded Python script using the python -c flag. This script is used to parse the local skills-config.yaml file. While the execution is dynamic, the source is provided within the skill and includes security hardening such as scrubbing sys.path to prevent module hijacking.
[COMMAND_EXECUTION]: The skill workflow requires the agent to execute several shell commands, including git worktree for environment isolation, npm install for dependency management, and gh pr create for pull request automation. These capabilities are within the scope of the skill's purpose but represent a local command execution surface.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external data sources to generate its output, which could be exploited via malicious content in those documents.
Ingestion points: skills-config.yaml, PRD shard files in the configured PRD_ROOT, and architecture documents in ARCH_ROOT.
Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore' instructions when processing the content of these external files.
Capability inventory: The agent has the ability to write new story files, update existing epic documents, and execute Git and CLI tool commands.
Sanitization: The read-config.sh script implements a strict YAML loader (_StrictLoader) to validate configuration data, preventing duplicate keys and ensuring the file mapping is well-formed.

create-parallel-stories