The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes project-level files to determine execution flow. It lacks explicit boundaries between data and instructions when reading project state.
Ingestion points: The skill reads 00_project/HANDOFF.md and 00_project/pipeline_progress.json to initialize or resume workflows.
Boundary markers: There are no specific delimiters or warnings used when reading these files to prevent the agent from following embedded instructions.
Capability inventory: The skill uses python scripts/pipeline_progress.py to perform state updates and file operations.
Sanitization: No sanitization or validation logic is described for the content ingested from the project files.
[COMMAND_EXECUTION]: The skill's primary mechanism of action involves executing a local Python script (scripts/pipeline_progress.py) with various arguments to manage project progress and checkpoints. This is the intended behavior for the skill's workflow management purpose.

pipeline-progress