Install
$ agentstack add skill-mike007jd-game-superpowers-gsp-subagent-build-loop ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Game Subagent Build Loop
Goal
Turn a game build into a default multi-agent workflow: builder, reviewer, verifier. The goal is stronger and more stable output from one user prompt, not thrift.
Outputs
Follow the gsp-orchestrator output strategy:
- inline (default): present subagent build loop plan in conversation.
- minimal or full: write
docs/game-studio/subagent-build-loop.md.
Default loop
For each meaningful build task or end-to-end slice:
- Dispatch a builder subagent to implement the slice.
- Dispatch a reviewer subagent using
gsp-build-review. - Dispatch a verifier subagent using
gsp-playability-verifier. - If review or verification fails, send the builder back with the concrete findings.
- Repeat until the slice is acceptable or the blocker should be surfaced to the user.
Parallel policy
Parallel builders are allowed only when ownership is disjoint. Examples: core loop vs HUD, feedback vs content pass, gameplay slice vs menus. Do not parallelize builders that fight over the same files or architectural choices.
Quality policy
- benchmark and showcase builds should prefer this loop by default
- greenfield polished-prototype work should usually prefer this loop
- live-risky work may still use a tighter reviewer / verifier loop with smaller tasks
Important
Do not claim completion after the builder finishes. Review and verification are expected to find problems.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: mike007jd
- Source: mike007jd/game-superpowers
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.