메뉴
Send Claude plan to an LLM for external technical review
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Formulate a single testable hypothesis by scanning the project for untested behaviors, gaps, and claims
Guided hypothesis-driven experimentation. Choose a project, generate hypotheses, select which to test, run experiments in parallel or sequentially, document findings.
Generate iteratively-reviewed research ideas from a problem statement. Fully guided - no arguments required.
Create research document with problem statement and context from multiple sources (repositories, papers, files)
Populate FINDINGS.md with results and analysis, update HYPOTHESIS.md status
Execute experiment run.sh, run analysis script, parse output, and interpret results
| name | review-plan |
| description | Send Claude plan to an LLM for external technical review |
| user-invocable | true |
| allowed-tools | ["Bash(**/review.sh *)","Bash(python3 *)","Bash(curl *)","Bash(jq *)","Glob"] |
| argument-hint | <plan_path> [model] [--dry-run] |
Get independent technical feedback on your Claude Code plan from an LLM.
IMPORTANT: Before running any commands, locate the review.sh script:
Glob: **/skills/review-plan/scripts/review.sh
Store the result as [REVIEW_SCRIPT]. Use this path for all subsequent bash commands.
If the script is not found, tell the user:
Error: Could not find review.sh script.
Make sure you're running from the sdlc-plugins directory or that the plugin is properly installed.
See README.md in this directory for setup instructions, available models, and troubleshooting.
/review-plan # auto-detect plan, use Azure/gpt-4o
/review-plan GCP/gemini-2.5-flash # use Gemini
/review-plan aws/claude-opus-4-6 # use Claude
/review-plan ~/.claude/plans/my-plan.md # explicit path
/review-plan --dry-run # verify config
.md → treated as plan fileprovider/model. Default: Azure/gpt-4o. See README.md for available models.--dry-run, --check-models, --help, --no-redactTest if models are reachable before running a full review:
/review-plan --check-models # test all 3 models
/review-plan --check-models Azure/gpt-4o # test single model
/review-plan --check-models aws/claude-opus-4-6 Azure/gpt-4o # test specific models
After finding [REVIEW_SCRIPT] using Glob, run:
[REVIEW_SCRIPT] [arguments]
Examples:
# Check model connectivity
[REVIEW_SCRIPT] --check-models aws/claude-opus-4-6 Azure/gpt-4o GCP/gemini-2.5-flash
# Review a plan
[REVIEW_SCRIPT] /path/to/plan.md Azure/gpt-4o
# Dry run
[REVIEW_SCRIPT] --dry-run