The Agent Skills Directory

[SAFE]: The skill primarily serves as a knowledge base and troubleshooting guide for a specific software platform. Analysis of the instructions and reference documents shows no signs of malicious intent, credential harvesting, or unauthorized network activity.
[PROMPT_INJECTION]: The skill implements a learning feedback loop by reading from and writing to references/learnings.md.
Ingestion points: Accumulated knowledge is read from references/learnings.md at the start of the interaction.
Boundary markers: Absent; the skill relies on the structured nature of the learning file.
Capability inventory: The skill provides textual advice, troubleshooting steps, and routing to other internal sales skills. It does not possess capabilities for arbitrary command execution or system modification.
Sanitization: Absent; the agent is instructed to append findings directly. This is a standard self-improvement pattern for informational agents and does not pose a significant risk given the restricted capabilities of the skill.
[COMMAND_EXECUTION]: While the skill mentions an npx command in the examples section, this is provided as a setup instruction for the user and is not executed by the agent itself. No automated or hidden command execution patterns were detected.

sales-sonnet