メニュー
Use when starting any conversation involving Codegen cloud agents — establishes skill map, decision tree, and integration with superpowers plugin
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | using-codegen-bridge |
| description | Use when starting any conversation involving Codegen cloud agents — establishes skill map, decision tree, and integration with superpowers plugin |
| user-invocable | false |
Codegen Bridge delegates coding tasks to cloud AI agents on the Codegen platform. Instead of making changes locally, tasks run in isolated cloud sandboxes and produce pull requests on GitHub.
When working with Codegen agents, invoke the relevant skill BEFORE taking action:
| Situation | Skill | How |
|---|---|---|
| Delegate a task to cloud agent | codegen-bridge:codegen-delegation | Auto-triggers on codegen_create_run |
| Execute a plan via cloud agents | codegen-bridge:executing-via-codegen | User-invocable or auto |
| Batch-delegate independent tasks | codegen-bridge:bulk-delegation | User-invocable |
| Monitor running agents | codegen-bridge:agent-monitoring | Auto-triggers on codegen_get_run |
| Debug a failed/stuck agent run | codegen-bridge:debugging-failed-runs | Auto-triggers on failure patterns |
| Write effective agent prompts | codegen-bridge:prompt-crafting | Auto-triggers on codegen_create_run |
| Review agent-created code/PRs | codegen-bridge:reviewing-agent-output | Auto-triggers on completed runs |
| Manage PR state (close/draft/ban) | codegen-bridge:pr-management | Auto-triggers on codegen_edit_pr |
| Analyze agent performance | codegen-bridge:run-analytics | User-invocable |
User wants to change code
├── Small/quick change (one-liner, config tweak)?
│ └── Do it locally — Codegen is overkill
├── Needs local environment (secrets, services, debugging)?
│ └── Do it locally
├── Feature implementation, refactor, test suite?
│ ├── Single task?
│ │ └── codegen-delegation → codegen_create_run
│ ├── Multi-step plan with dependencies?
│ │ └── executing-via-codegen → sequential runs
│ └── Multiple independent tasks?
│ └── bulk-delegation → codegen_bulk_create_runs
└── Not sure?
└── Ask: "Should this run locally or via Codegen cloud agent?"
If the superpowers plugin is installed alongside codegen-bridge:
Use superpowers for PROCESS:
superpowers:brainstorming — design before codesuperpowers:writing-plans — create implementation planssuperpowers:test-driven-development — TDD disciplinesuperpowers:systematic-debugging — debug LOCAL issuessuperpowers:requesting-code-review — review LOCAL codesuperpowers:verification-before-completion — verify LOCAL workUse codegen-bridge for CLOUD EXECUTION:
codegen-bridge:executing-via-codegen — execute plans via cloud agentscodegen-bridge:debugging-failed-runs — debug AGENT failurescodegen-bridge:reviewing-agent-output — review AGENT-created codecodegen-bridge:prompt-crafting — optimize prompts for agentsCombined workflow:
superpowers:brainstorming → design the featuresuperpowers:writing-plans → create the plancodegen-bridge:executing-via-codegen → execute via cloud agentscodegen-bridge:reviewing-agent-output → review agent PRssuperpowers:finishing-a-development-branch → mergeIf superpowers is NOT installed, codegen-bridge works standalone. You lose the process skills but all cloud delegation skills work independently.
| Command | Purpose |
|---|---|
/codegen | Delegate a task to cloud agent |
/cg-status | Check status of agent runs |
/cg-logs | View agent execution logs |
/cg-merge | Merge PRs from agent runs |
/cg-settings | Manage plugin settings |
Use when batch-delegating multiple independent tasks to Codegen cloud agents in a single call. Uses codegen_bulk_create_runs for parallel execution. Best for plans where tasks have no inter-dependencies.
Use when debugging failed, stuck, or misbehaving Codegen agent runs — systematic 4-phase approach. Triggers on agent run failures, timeouts, or unexpected results. Analogous to superpowers systematic-debugging but for cloud agents.
Use when reviewing code and PRs created by Codegen agents — two-stage review process. Stage 1 checks spec compliance (did the agent do what was asked?). Stage 2 checks code quality (is the code good?). Triggers after agent runs complete.
Use when analyzing Codegen agent run performance — interpret success rates, duration metrics, token usage, and trends via codegen_get_run_analytics. Best used after batch runs or for periodic performance review.
Guide for writing effective prompts for Codegen cloud agents. Includes templates, anti-patterns, and optimization strategies. Auto-triggers when creating agent runs to ensure prompt quality.
Monitor running Codegen agents — poll status, review logs, detect blockers, and surface progress. Triggered when codegen_get_run, codegen_get_logs, codegen_list_runs, or codegen_get_execution_context tools are invoked.