The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from an external file, .claude/product-marketing-context.md, to gain context about the user's product and audience.
Ingestion points: Defined in SKILL.md, where the agent is instructed to read the marketing context file if it exists.
Boundary markers: No specific delimiters or "ignore" instructions are provided for the ingested content.
Capability inventory: The skill possesses the capability to execute the local Python script scripts/tool_roi_estimator.py. It does not have network access or file-writing capabilities described in the skill files.
Sanitization: No sanitization or validation of the ingested context file is mentioned.
[DYNAMIC_EXECUTION]: The skill includes and utilizes a local Python script, scripts/tool_roi_estimator.py, for modeling the return on investment (ROI) of building marketing tools.
Evidence: SKILL.md references the execution of the script for ROI modeling. The script itself uses standard Python libraries for calculations and JSON processing without using unsafe dynamic execution functions like eval() or exec().

free-tool-strategy