counter-narrative

Warn

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes a local script named narrative-mapper.py with the generate-counter argument to select a strategy based on competitor analysis.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its processing of external, untrusted competitor data.
  • Ingestion points: Competitor evidence such as URLs, landing pages, and messaging artifacts described in the process section.
  • Boundary markers: None; there are no instructions for the agent to ignore or delimit instructions contained within analyzed competitor materials.
  • Capability inventory: The skill reads brand configuration files from the ~/.claude-marketing/ directory and executes a Python script via shell command.
  • Sanitization: None; the skill examines evidence to determine positioning intent and strategy without a verification or sanitization step for the external content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 9, 2026, 12:16 AM