The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Git commands (git status) to freeze state and executes local Python scripts within the project structure to handle task reconciliation and reporting.
[DYNAMIC_EXECUTION]: The instructions direct the agent to execute specific Python scripts including workflow/scripts/open_items.py, workflow/scripts/record.py, and codebase-map/scripts/task_report.py to automate the generation of delivery outcomes.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from requirement tickets, acceptance items, and review findings to perform its reconciliation logic, which could contain malicious instructions.
Ingestion points: Reads requirement tickets, actor lines, and review.json files within the workflow/ directory.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present when reading external ticket or review data.
Capability inventory: Includes file system read/write access and the ability to execute local Python scripts.
Sanitization: The skill does not define validation or sanitization procedures for the content extracted from tickets or reviews.

workflow-deliver