The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources, which creates a potential surface for indirect prompt injection.
Ingestion points: The instructions direct the agent to "Read any files [the user] provide" and "Search the web for the market, competitors... and read any sources the user points to" (SKILL.md).
Boundary markers: There are no explicit instructions to use delimiters (like XML tags or triple quotes) or to provide the agent with "ignore embedded instructions" warnings for the external content it reads.
Capability inventory: The skill possesses the capability to search the web, read files, and write the output to a new markdown document (PRD-[product-name].md).
Sanitization: The instructions do not specify any sanitization, filtering, or validation of the external content before it is interpolated into the agent's context.
[SAFE]: The external links provided in the "Further Reading" section point to productcompass.pm, which is a professional blog site associated with the skill author's methodology. These links are for reference purposes and do not represent a security risk.
[SAFE]: No instances of hardcoded credentials, malicious command execution, or unauthorized data exfiltration were identified within the skill's instructions.

create-prd