The Agent Skills Directory

[SAFE]: The skill utilizes well-known job search platforms and official government portals for its operations, including job.mohrss.gov.cn (Ministry of Human Resources and Social Security), iguopin.com, and zhipin.com. These are legitimate services appropriate for the skill's intended purpose.
[INDIRECT_PROMPT_INJECTION]: The skill's architecture involves ingesting data from external websites, which presents a surface for indirect prompt injection if a job posting contains adversarial instructions.
Ingestion points: Untrusted data enters the agent context through the use of the web_fetch tool when the agent retrieves details from job announcement pages as described in the search process of SKILL.md.
Boundary markers: Absent. The skill does not define specific delimiters or instructions to ignore embedded prompts within the fetched web content.
Capability inventory: The skill's capabilities are limited to information extraction, summarization, and reporting. It does not have access to sensitive local files, shell command execution, or file writing tools.
Sanitization: No explicit sanitization or validation of the fetched HTML/text content is specified in the instructions; the agent relies on its internal reasoning to extract specific fields like salary and recruitment dates.

taxue-job-search