paperclip-converting-plans-to-tasks

Overview

paperclip-converting-plans-to-tasks is a Claude Code skill from the Paperclip project that provides industry-agnostic guidance on how to translate any plan into properly structured, executable Paperclip issues. It does not prescribe a plan format. Instead, it teaches you how to take whatever plan you have and convert it into assigned issues with the right agent specialty, dependency edges (blockedByIssueIds), and parallelization so that Paperclip's executor can automatically pick up and sequence the work. This skill is a companion to the core paperclip skill, which covers the API mechanics of recording plans and reassigning issues.

Key Features

• Specialty-based assignment: Emphasizes looking up the company's available agents and their specialties before assigning work, ensuring each piece goes to the most qualified agent rather than defaulting everything to yourself
• Dependency tree wiring: All hard dependencies must be expressed through blockedByIssueIds on the dependent issue, not as prose. When a blocker reaches "done", dependents automatically wake, enabling Paperclip's executor to manage sequencing without manual intervention
• Parallel-first sequencing: Independent branches of the dependency graph should start in parallel. Unlike humans, agents allow concurrent runs, so parallel work can be assigned to the same agent when appropriate
• Deep planning over shallow: Encourages capturing real detail including goals, constraints, unknowns, success criteria, and risks. A shallow plan becomes rework downstream because assignees can only act on what they can read
• Anti-procrastination guardrails: Includes explicit guidance against over-planning. If the next step is small and clear, just do it. Re-planning a plan or splitting work that one agent could finish faster than the breakdown takes is recognized as procrastination

Who is this for?

• Paperclip agent operators who need to break complex projects into well-structured, dependency-aware task graphs that the executor can automatically process
• AI agents acting as project managers or planners within a Paperclip company who need disciplined guidance on task decomposition and assignment
• Teams adopting Paperclip who want to understand best practices for converting high-level plans into actionable, parallelized work items