lightrun-runtime-aware-pr-review

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh api) to programmatically fetch pull request details and diff data. These operations target the official GitHub API and are fundamental to the skill's purpose of code review.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external content from GitHub pull request diffs and captures production runtime data via the Lightrun MCP.
  • Ingestion points: Pull request diffs (via gh api) and variable values captured from production environments (via lightrun MCP tools).
  • Boundary markers: The process is governed by a structured multi-phase workflow defined in the skill's reference files.
  • Capability inventory: Includes the ability to interact with the GitHub API and production runtime through snapshots.
  • Sanitization: There are explicit, mandatory instructions to sanitize all captured runtime evidence, ensuring that secrets, tokens, and identifiers are redacted before being included in the final report.
  • [SAFE]: All external communications, including instrumentation and telemetry, are directed to the vendor's official domain (lightrun.com). The skill explicitly prohibits state-changing instrumentation and mandates strict data sanitization, minimizing risks associated with production data access.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 11:20 AM
Security Audit — agent-trust-hub — lightrun-runtime-aware-pr-review