The Agent Skills Directory

[SAFE]: The skill implements logic for analyzing marketing performance metrics (CTR, CVR, ROAS, CPA) using a provided CSV structure and a Python script.
[DATA_EXF_REMOVED]: No data exfiltration or unauthorized network communication was detected. The script and instructions focus solely on local data processing and reporting.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns, such as fetching and piping scripts to a shell, were found. The included Python script uses standard libraries for mathematical calculations.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data (CSV) and generates natural language interpretations. This presents a potential surface for indirect prompt injection if malicious instructions are embedded in fields like 'campaign_name' or 'segment'.
Ingestion points: CSV input data defined in SKILL.md.
Boundary markers: Absent; data is processed directly into metrics and interpretations.
Capability inventory: The agent generates summaries and recommends budget changes based on logic in references/budget_reallocation_rules.md.
Sanitization: No explicit sanitization or filtering of CSV string content is defined before the agent interprets the results.

analyzing-marketing-campaign