The Agent Skills Directory

[SAFE]: The skill functions as a conductor or orchestrator, routing user queries to a defined set of investor and research thinker-lens skills. No malicious code, unauthorized command execution, or data exfiltration patterns were detected.
[COMMAND_EXECUTION]: The skill references the use of external platform tools such as analyse-smartmoney-polymarket and feed-fomc to ground economic debates in market data. These are legitimate tool invocations consistent with the skill's primary purpose.
[PROMPT_INJECTION]: The instructions are focused on ensuring intellectual honesty and preserving the unique frameworks of each panelist. There are no patterns suggesting attempts to override safety guardrails or bypass system instructions.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external inputs (user questions, market odds, and Fed statements). While this creates an attack surface for indirect injection, the instructions provide a structured framework for analysis which helps mitigate accidental obedience to embedded instructions. The risk is evaluated as low and typical for analytical skills.
[CREDENTIALS_SAFE]: No hardcoded credentials, API keys, or sensitive file paths were found in the skill or its source manifest.

macro-panel