The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external bid data (JSON or text) which can be manipulated by third parties to influence the agent's behavior through reflected instructions.
Ingestion points: The scripts/run.mjs script reads user-supplied files specified by the --input flag.
Boundary markers: No explicit delimiters or instructions to ignore embedded content are used when processing the data.
Capability inventory: The skill is configured with Read and Bash tool permissions.
Sanitization: Although the engine performs string normalization (removing whitespace and punctuation), it does not filter or sanitize natural language instructions embedded within the bid item names.
[EXTERNAL_DOWNLOADS]: The documentation contains instructions to install a package from a well-known service provider as part of the upgrade process.
Evidence: The SKILL.md file instructs the agent to run npx -y @alipay/agent-payment@latest install to enable purchase functionality for the full version.
Context: The package originates from Alipay, a well-known technology service, and is part of the documented installation flow for the vendor's commercial offering.

bid-comparison-free