The Agent Skills Directory

[SAFE]: The skill is designed for stock market research and uses standard tools like web_search and delegate_task to achieve its goals. There is no evidence of credential theft, hardcoded secrets, or unauthorized network exfiltration.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests data from external web searches to drive its internal logic and subagent tasks.
Ingestion points: Web search results for stock prices, company overviews, news headlines, and financial metrics are gathered in SKILL.md (Phase 1) and compiled into a 'DISCOVERY_BRIEF'.
Boundary markers: Absent. The search results are formatted into a brief and passed to subagents without explicit delimiters or instructions to ignore embedded commands within the search content.
Capability inventory: The skill possesses the ability to perform further web searches, delegate tasks to other agents, and write markdown files to the local file system.
Sanitization: Absent. The skill does not perform any validation or filtering on the content returned by the search engine before incorporating it into the agent's context.

trade-analyze