The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources such as llms.txt files and search API responses on the aguasto.com domain. While this constitutes an attack surface, the instructions specifically guide the agent to distinguish between announced and measured capacities and to preserve original units, which promotes accurate data handling.
[DATA_EXFILTRATION]: The skill performs GET and POST network requests to retrieve information from a specific domain. These operations are transparently defined and align with the skill's stated purpose of hardware research. No access to sensitive local environment variables or private files is requested or performed.
[SAFE]: The skill contains no obfuscation, privilege escalation attempts, persistence mechanisms, or remote code execution patterns. The logic is straightforward and consistent with its documentation.

aguasto-research