speckit-code-review

Pass

Audited by Gen Agent Trust Hub on Nov 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the bash tool to execute a wide range of shell commands for repository analysis and test execution. This includes git for diffing and submodule expansion, as well as language-specific test runners such as npx jest, npx vitest, pytest, go test, mvn, gradlew, and dotnet test. These commands are used to gather data for the review process and calculate test coverage.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a significant attack surface for indirect prompt injection because its core function is to ingest and analyze untrusted external data, specifically feature specifications (spec.md) and project source code.
  • Ingestion points: The agent reads content from specs/<feature-folder>/spec.md and all files identified in the git change set (staged and unstaged).
  • Boundary markers: The instructions do not define clear delimiters or "ignore" instructions to prevent the agent from following natural language commands that might be embedded within the specifications or code comments being reviewed.
  • Capability inventory: The agent performing the review has access to powerful tools including bash (for shell execution), view/read_file, create/write_file, and edit/patch, which could be abused if an injection is successful.
  • Sanitization: There is no evidence of sanitization, escaping, or filtering of the spec or code content before it is processed by the LLM for review.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 2, 2026, 05:37 AM