The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The aeo_audit.py script uses urllib.request to fetch content from arbitrary URLs provided by the user. There are no restrictions or sanitization for destination IP addresses or domains.
Evidence: urllib.request.urlopen in scripts/aeo_audit.py.
[COMMAND_EXECUTION]: The skill requires the execution of multiple Python scripts (aeo_audit.py, aeo_optimizer.py, citation_tracker.py) which interact with the local file system and network.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external URLs and local files to generate reports and optimized content. This content is then returned to the agent's context, creating a potential surface for indirect prompt injection if the source content contains malicious instructions.
Ingestion points: aeo_audit.py fetches data from --url or --input file; aeo_optimizer.py reads from --input file.
Boundary markers: None present in the script outputs to delimit untrusted content.
Capability inventory: Local file read/write via pathlib, network access via urllib.
Sanitization: No sanitization or escaping of the ingested content is performed before it is processed by the signal detection logic.

aeo