kafka-dlq-review

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it ingests and processes untrusted data from Kafka DLQ messages to verify metadata completeness. \n
  • Ingestion points: The execute_sql tool output in SKILL.md (Step 4), which samples messages from discovered DLQ topics. \n
  • Boundary markers: Absent; the instructions do not specify any delimiters or ignore-instructions for the sampled message content. \n
  • Capability inventory: The skill uses Bash, Grep, Glob, Read, and various Lenses MCP tools to perform its audit and report results. \n
  • Sanitization: Absent; no sanitization or escaping of the sampled Kafka message data is described. \n- [COMMAND_EXECUTION]: The skill utilizes Bash, Grep, and Glob (Step 6) to scan the local codebase for error handling patterns and DLQ implementation details. While these are powerful tools, their use here is aligned with the skill's primary purpose of conducting a thorough DLQ review.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 12:44 PM