competitive-teardown
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze data from untrusted external sources, including competitor websites, user reviews (e.g., from G2 or Capterra), and social media platforms. This creates a surface for indirect prompt injection, where an attacker could embed malicious instructions in public content to manipulate the AI agent's analysis or recommendations.
- Ingestion points: External data sources identified in
SKILL.md, such as review sites, competitors' homepages, and job boards. - Boundary markers: The skill instructions do not specify the use of delimiters or explicit 'ignore' directives to sanitize or isolate the collected data before it is processed by the agent.
- Capability inventory: The provided Python scripts (
scripts/battle_card_generator.py,scripts/competitor_scorer.py, andscripts/feature_matrix_builder.py) are focused on data analysis and lack harmful capabilities such as network access, file system modifications, or arbitrary code execution. - Sanitization: No data validation or sanitization is performed on the ingested content within the provided scripts.
Audit Metadata