The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's workflow involves processing user feedback and growth metrics in Step 2, which introduces a surface for indirect prompt injection if that data contains malicious instructions designed to subvert the agent's behavior.
Ingestion points: Step 2 (/pm-report) processes data reports and user feedback analysis.
Boundary markers: The instructions do not define explicit boundary markers or delimiters to isolate untrusted user feedback from the agent's system instructions.
Capability inventory: The skill has access to powerful tools including Bash, Write, Read, and WebSearch across its workflow steps.
Sanitization: There is no mention of sanitization, filtering, or validation of the input data before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill instructions provide a shell command to install dependencies from an external repository located at https://github.com/konglong87/superPM. This repository is owned by the skill's author.

analyze-growth