The Agent Skills Directory

Prompt Injection (SAFE): The skill documentation and command-line instructions contain no attempts to override system prompts, bypass safety filters, or extract system instructions.
Data Exposure & Exfiltration (SAFE): No hardcoded credentials, API keys, or sensitive file path exposures (e.g., .ssh or .aws) were found. Logging is directed to a local directory defined in the environment's repository structure.
Unverifiable Dependencies & Remote Code Execution (SAFE): The scripts rely exclusively on Python standard libraries (argparse, json, math, pathlib). There are no external package installations or downloads from remote sources.
Indirect Prompt Injection (SAFE): A surface for indirect prompt injection exists in scripts/optimize_conversion_funnel.py, which reads from a user-supplied JSON file. However, the risk is categorized as SAFE as the script performs only mathematical analysis and lacks the capability to execute commands or make network requests based on the data.
Ingestion points: scripts/optimize_conversion_funnel.py (via --funnel-data argument).
Boundary markers: Absent.
Capability inventory: The script calculates conversion statistics and outputs recommendations; no subprocess or network calls are present.
Sanitization: Uses standard json.load() for structured parsing.
Dynamic Execution (SAFE): scripts/utils.py dynamically appends the repository root to sys.path to enable local imports. This is a standard project utility and does not involve executing untrusted strings or loading remote modules.

monetization-strategist