The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to interact with a vendor-specific CLI (postplus). Specifically, it uses postplus quote confirm --json --challenge-file <challenge.json> to handle quote confirmations. This is a functional requirement of the skill and uses vendor-owned tooling.
[INDIRECT_PROMPT_INJECTION]: The skill functions by analyzing potentially untrusted data from external sources.
Ingestion points: Marketplace exports, supplier sheets, and quotation data (SKILL.md).
Boundary markers: Absent; the instructions do not specify delimiters or instructions to ignore data-embedded commands.
Capability inventory: Subprocess execution of the postplus CLI and invocation of research modules like google-trends-research and tiktok-research (SKILL.md).
Sanitization: Absent; there are no instructions regarding validation or escaping of ingested data.

sourcing-selection