The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions intended to influence the agent's behavior.
Ingestion points: The skill ingests a user-provided plan via the <plan> argument and processes external data files (sources.json, profile.json, corpus.json) using Python scripts as described in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or treat embedded instructions within the processed files as data only.
Capability inventory: The skill has the capability to execute shell commands and run local Python scripts (ai_training_data_audit.py, data_product_strategy_picker.py, data_asset_valuator.py) which process the input data.
Sanitization: No sanitization, escaping, or validation steps for the external content are specified in the skill instructions.
[COMMAND_EXECUTION]: The skill workflow involves the execution of shell commands to run local Python scripts.
Evidence: The Workflow section in SKILL.md contains multiple python command execution patterns targeting scripts in relative paths (e.g., ../../../c-level-advisor/skills/chief-data-officer-advisor/scripts/ai_training_data_audit.py).

cdo-review