The Agent Skills Directory

[SAFE]: Analysis of the skill instructions and reference documents reveals no malicious intent. The focus is entirely on a legitimate financial analysis use case with robust procedures for data verification and citation.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and process data from external, potentially attacker-controlled sources such as web search results, earnings call transcripts, and investor relations pages. This is an inherent risk for agents performing web-based data retrieval.
Ingestion points: Web search queries for company earnings, investor relations news sections, and transcript repositories (referenced in SKILL.md and references/workflow.md).
Boundary markers: The skill does not define specific prompt delimiters but includes rigorous verification steps (Phase 1, Step 1c) requiring the agent to manually check release dates and cross-reference multiple documents to ensure the data is current and authentic.
Capability inventory: Python execution (matplotlib, pandas, seaborn) for data plotting and a DOCX generation tool for report output.
Sanitization: No explicit sanitization or filtering of external text content is specified in the instructions.
[EXTERNAL_DOWNLOADS]: The skill refers to well-known financial services and official government repositories for data acquisition.
Evidence: Instructions direct the agent to use SEC EDGAR (sec.gov) and reputable financial news sites like Seeking Alpha for data gathering. These are well-known technology and government services and do not escalate the security verdict.

earnings-analysis