From clp
Plan a task with context budget awareness. Breaks work into context-sized chunks, estimates token cost per phase, and suggests when to checkpoint or use subagents. This skill should be used before starting any non-trivial task, when the user asks to "plan this", "estimate tokens", or "break this into phases".
How this skill is triggered — by the user, by Claude, or both
Slash command
/clp:clp-planThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
<context>
<planning_steps> Check current context utilization with /clp:status. Calculate available working tokens for this task.
Break the task into phases that each fit within a single context window. Each phase should be independently completable — if compaction fires between phases, the handoff manifest should be sufficient to continue. Which phases involve heavy file reading or exploration? These should be delegated to subagents to keep the main context clean. Mark these as "subagent: true" in the plan. For each phase, estimate token cost: - File reads: ~100 tokens per KB of source code - Tool outputs: ~500-2000 tokens per bash command - Conversation: ~200 tokens per turn (user + assistant) - Skill loading: check estimated_tokens in skill-registry.json Insert /clp:checkpoint calls at natural breakpoints: - After each phase completes - Before any phase that might trigger compaction - Before switching from planning to implementation<plan_format>
Available budget: [X]k tokens remaining Estimated total cost: [Y]k tokens Phases: [N] | Checkpoints: [M] | Subagent tasks: [P]
...
npx claudepluginhub ryannle1/clp --plugin clpAssesses codebase to classify task complexity and generates scaled plans—from quick outlines to full design docs—for features, refactors, or multi-step tasks before coding.
Creates structured implementation plans for multi-step tasks from specs or requirements. Generates plan.json and task markdown files with explicit code and testing steps.
Generates executable Markdown implementation plans for multi-step tasks from context briefs, resolving ambiguities, ordering dependencies, and enabling parallel worker execution.