analyst-wkh

Provides structured US equity research including company snapshot, earnings analysis, valuation scenarios, catalyst/risk assessment, and actionable monitorin...

This skill is internally coherent and low-risk as an instruction-only template, but consider these practical points before installing: (1) The skill relies on either user-supplied data or platform tooling (get_price, get_fundamentals, etc.) to fetch live financial data — verify the trustworthiness and permissions of those tool implementations on your platform. (2) Because the skill is only instructions, an agent without data access will produce outputs based on model knowledge and assumptions; insist on source citations and supply primary data (filings, transcripts, price history) for numerical work. (3) Do not paste confidential or proprietary documents unless you are comfortable sharing them with the agent. (4) The skill includes investment recommendations structure but is not a substitute for professional advice — verify valuations and trade plans independently. (5) If you want to prevent autonomous runs, consider disabling model invocation for untrusted skills via your platform settings. Overall this skill appears consistent with its stated purpose.

Type: OpenClaw Skill Name: us-stock-analyst-wkh Version: 1.0.2 The skill bundle describes a legitimate US stock analyst function. The `skill.md` file provides clear instructions for the AI agent, including guardrails against fabricating data and enforcing a structured output. The `system_prompt` actively constrains the agent to be factual and follow the defined structure, which mitigates prompt injection risks. The listed tools (`get_price`, `get_fundamentals`, etc.) are standard for financial analysis and are invoked without any instructions for malicious input crafting or unauthorized actions. There is no evidence of intentional harmful behavior, data exfiltration, malicious execution, or persistence mechanisms within the provided files.