Parallel Agent Contracts

When you're running parallel Claude agents to implement code, they'll happily recreate the same types in different files, breaking your build. This skill gives you a prompt template that forces each agent to grep for existing types before creating new ones, includes a canonical type ownership map (like NormalizedTool lives in agent.ts, Message lives in types.ts), and makes them run tsc --noEmit before marking work complete. It's basically a contract system enforced by TypeScript's compiler rather than hoping agents remember context. Worth using if you're spawning multiple agents to work on a TypeScript codebase simultaneously and tired of merge conflicts from duplicate interface definitions.

skills.sh installs: 480
GitHub stars: 3.9K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add parcadei/continuous-claude-v3 --skill parallel-agent-contracts --agent claude-code

Installs into .claude/skills of the current project.

Files

Parallel Agent Type Contracts

When launching parallel agents for code implementation, prevent type duplication.

Required in Every Agent Prompt

1. Verification Command (MANDATORY)

## Before Marking Complete
Run verification:
\`\`\`bash
npx tsc --noEmit 2>&1 | head -20
\`\`\`
If ANY type errors exist, fix them before completing.

2. Grep-Before-Create

## Before Creating Any Type/Interface
First check if it exists:
\`\`\`bash
grep -r "interface YourTypeName\|type YourTypeName" src/
\`\`\`
If found, import it. NEVER duplicate existing types.