Cninfo Announcement Scraper

Fetch and filter CNINFO official disclosures by event and industry to extract positive catalysts for A-share monitoring with structured output.

Security Scan

The skill's declared purpose (scraping CNINFO disclosures and extracting catalysts) aligns with its requirements and instructions, but the SKILL.md is high-level and provenance is unclear, so you should verify how the agent will fetch data and which endpoints are allowed before installing.

Purpose & Capability

Name and runtime instructions all describe pulling and classifying CNINFO announcements; there are no unrelated required binaries, credentials, or config paths requested. The requested capabilities are proportionate to the stated goal.

Instruction Scope

SKILL.md stays within the task (fetch, parse, classify, filter, return structured rows) and explicitly warns to use official disclosure links only and not to provide auto-trade instructions. However, the instructions are high-level and do not specify exact endpoints, APIs, or whitelisted hosts — leaving the agent discretion about how/where to fetch data, which could result in calls to unexpected external services if not constrained.

Install Mechanism

Instruction-only skill with no install spec and no code files — minimal on-disk risk. Nothing is downloaded or written by an installer.

Credentials

The skill declares no environment variables, credentials, or config paths. This is proportionate to a read-only scraping/classification task.

Persistence & Privilege

always is false and there are no indications the skill requests permanent system-level presence or modification of other skills. Agent-autonomous invocation is allowed by platform default but not unique to this skill.

Assessment

This skill appears coherent for extracting CNINFO disclosures, but before installing: (1) verify the agent will only fetch from official CNINFO endpoints (supply explicit API/URL whitelist if possible), because the SKILL.md is high-level and leaves scraping method unspecified; (2) ensure the agent has only read-only network access and no permissions to execute trades or access unrelated credentials; (3) confirm the source/author (none provided) or test in a sandbox to observe which external endpoints are contacted; (4) check legal/terms-of-service implications for scraping CNINFO; and (5) if you want tighter safety, ask the skill author to list exact endpoints, response examples, and a sample structured output so you can validate behavior before granting broader access.

Use this skill to pull CNINFO official disclosures and extract positive catalysts for A-share monitoring.

Inputs

• Time window (default: current trading day)
• Event filter (default: 业绩预增, 重大合同, 政策扶持)
• Industry filter (default: 新能源, 消费)

Procedure

1. Fetch latest CNINFO announcements.
2. Parse and normalize: code, company, title, announcement_time, url.
3. Classify announcement type.
4. Keep only allowed event types and industries.
5. Return structured rows for alert generation.

Output Schema

• ticker
• company
• catalyst
• source_url
• timestamp
• confidence
• trigger_price
• invalidation_condition

Safety

• Use official disclosure links only.
• Never provide auto-trade instructions.