The Agent Skills Directory

[DATA_EXFILTRATION]: The skill recommends using curl with https://defuddle.md/ to fetch external content. This workflow sends target URLs to a third-party service, which could result in the exposure of metadata or browsing intent to an external entity.
[PROMPT_INJECTION]: The skill provides instructions for the agent to fetch and process untrusted external data. Ingestion points: Results from curl requests to the defuddle.md service as described in SKILL.md. Boundary markers: Absent; the skill does not specify the use of delimiters or instructions to ignore embedded commands in fetched content. Capability inventory: The skill makes use of sigma conversion tools, atomic-red-team emulation commands, and execution of local Python scripts. Sanitization: None; the skill does not describe any validation or filtering of the content retrieved from the external service.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to the non-whitelisted utility domain defuddle.md to retrieve markdown content.
[COMMAND_EXECUTION]: The skill utilizes several CLI tools including sigma, atomic-red-team, and local Python scripts (scripts/sync_attack.py, scripts/rules_to_navigator.py) for its core functionality. While standard for detection engineering, these constitute an active command execution surface.

engineering-detections