tw-research-rebuttal-writer

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) due to its core functionality of processing external data.
  • Ingestion points: In SKILL.md (Steps 1, 3, and 5), the user is instructed to paste 'Reviewer Comments' directly into the conversation context.
  • Boundary markers: The skill lacks explicit delimiters or specific 'ignore-instructions' directives to protect the agent from adversarial commands embedded within the pasted reviewer text.
  • Capability inventory: The skill is configured with Bash, Read, and Write permissions in the allowed-tools frontmatter, making successful injection highly impactful.
  • Sanitization: There is no evidence of sanitization, escaping, or validation of the external content before it is interpolated into the prompt and categorized by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 04:11 AM
Security Audit — agent-trust-hub — tw-research-rebuttal-writer