The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external files, which could contain instructions meant to influence the agent's behavior.
Ingestion points: Reads project requirements from ai/memory-bank/site-setup.md and refers to detailed instructions in ai/agents/pm.md.
Boundary markers: The instructions lack explicit security delimiters or clear "ignore embedded instructions" warnings for the ingested content, though it does instruct the agent to "Quote EXACT requirements."
Capability inventory: The skill has the ability to write to the file system (ai/memory-bank/tasks/[project-slug]-tasklist.md) and execute a local shell script (./qa-playwright-capture.sh).
Sanitization: There is no mention of sanitizing or validating the content of the specification file before it is used to generate tasks or influence command execution.
[COMMAND_EXECUTION]: The skill instructions include a requirement to execute a local shell script for quality assurance.
Evidence: The template specifies the use of ./qa-playwright-capture.sh http://localhost:8000 public/qa-screenshots to perform testing. While the command appears static, its execution is triggered based on the processing of the external specification file.

Senior Project Manager