The Agent Skills Directory

[PROMPT_INJECTION]: Indirect prompt injection surface identified. * Ingestion points: EXAMPLES.md (Example 4) describes a workflow where the agent reads product information from external Excel files. * Boundary markers: Absent. No delimiters or specific instructions are provided to the agent to ignore or isolate potentially malicious instructions embedded within the Excel data. * Capability inventory: SKILL.md defines tools xianyu_publish_product and xianyu_batch_publish_products which perform authenticated network requests to publish content to an external platform. * Sanitization: Absent. There is no evidence of input validation, escaping, or filtering for data ingested from external sources before being processed by the tools.
[NO_CODE]: This skill is composed solely of markdown documentation and YAML configuration. No executable scripts, binaries, or source code files are included in the skill package.

xianyu-product