mcs-iterate Skill
Use this skill to autonomously build, evaluate, and fix an MCS agent until it reaches SHIP verdict. Composite of /mcs-build → /mcs-eval → /mcs-fix → re-eval, max 3 fix cycles per agent. Reuses the iterate-orchestrator audit + verdict primitives. Use after /mcs-research when agentspec.json is ready and the user wants th Published by microsoft in MCS-Agent-Builder.
What is mcs-iterate Skill?
Use this skill to autonomously build, evaluate, and fix an MCS agent until it reaches SHIP verdict. Composite of /mcs-build → /mcs-eval → /mcs-fix → re-eval, max 3 fix cycles per agent. Reuses the iterate-orchestrator audit + verdict primitives. Use after /mcs-research when agentspec.json is ready and the user wants th Published by microsoft in MCS-Agent-Builder. This profile combines repository metadata with install, compatibility, and usage signals so developers can quickly decide whether it fits their agent workflow before opening the source repository.
Automated repository signals based on public metadata such as recency, license, installation evidence, and adoption. These are not a security audit or endorsement.
Key capabilities
- Includes SKILL.md support
- Reusable instructions support
- Security review
- Research
- Security review use cases
- Research use cases
Technical details
- Install or run with Copy skill directory
When to use mcs-iterate Skill
- Use it for security review.
- Use it for research.
Built with
Editorial notes
Source
- Creator: microsoft
- Repository: microsoft/MCS-Agent-Builder
- Skill file: .claude/skills/mcs-iterate/SKILL.md
What it does
Use this skill to autonomously build, evaluate, and fix an MCS agent until it reaches SHIP verdict. Composite of /mcs-build → /mcs-eval → /mcs-fix → re-eval, max 3 fix cycles per agent. Reuses the iterate-orchestrator audit + verdict primitives. Use after /mcs-research when agentspec.json is ready and the user wants th
Skill instructions
/mcs-iterate — Autonomous MCS Build → Eval → Fix Loop (Layer B) Single command for fully-autonomous MCS agent shipping. Runs the full Build-Guides flow without manual phase transitions: build the agent, run evals, classify failures, apply fixes, re-eval, loop until SHIP or hard cap. This is the Layer B counterpart to /iterate. It does NOT touch Layer A code (frontend/backend/framework). It only iterates on the MCS agent state via the existing /mcs- skills, reusing the iterate-orchestrator's audit + verdict primitives where appropriate. When to fire - User typed /mcs-iterate {projectId} {agentId}. - User said "ship the agent", "iterate to SHIP", "build, eval, fix until ready". - After /mcs-research produced a complete agentspec.json and the user wants the agent shipped autonomously. - To resume a partial build where agentspec.json.buildStatus shows incomplete steps and evalConfig.lastVerdict is BLOCK or ITERATE. When NOT to fire - The agentspec.json is incomplete — run /mcs-research fir
Explore related resources
Frequently asked questions
What is mcs-iterate?
mcs-iterate is a open-source AI agent skill with Copy skill directory. Use this skill to autonomously build, evaluate, and fix an MCS agent until it reaches SHIP verdict. Composite of /mcs-build → /mcs-eval → /mcs-fix → re-eval, max 3 fix cycles per agent.
Who is mcs-iterate best for?
mcs-iterate is best for reusing agent instructions, scripts, and references, security review workflows, research workflows.
How do I install mcs-iterate?
Install or run mcs-iterate using Copy skill directory. Check mcs-iterate for the latest setup command.
Is mcs-iterate actively maintained?
mcs-iterate may need a closer maintenance check before production use.
Auto-fetched from GitHub.
