The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches data from an external API and user arguments to generate reports, creating a surface where the agent processes potentially untrusted content.\n
Ingestion points: Data enters the agent's context through the ContextStellar stats API response and the $ARGUMENTS variable in SKILL.md.\n
Boundary markers: The instructions lack explicit markers or warnings to disregard instructions embedded within the external metrics or analyzed prompts.\n
Capability inventory: The skill utilizes curl for network requests but does not possess high-privilege capabilities such as file system writing or arbitrary code execution.\n
Sanitization: There is no evidence of data sanitization or validation of the fetched API response before it is used for logic generation.\n- [COMMAND_EXECUTION]: The skill uses the curl tool to retrieve project metrics and learned weights.\n- [EXTERNAL_DOWNLOADS]: The skill fetches performance data and configuration from contextstellar.com to fulfill its primary purpose.

optimize