What problem does it solve? AI coding agents lose their working memory when the context window fills, a session is compacted, or /clear is run, causing repeated work and lost progress on multi-step tasks. This Skill stores the plan, findings, and progress in markdown files on disk so work can be resumed at any point. ## Core Features & Use Cases - Persistent planning files: Creates and maintains task_plan.md, findings.md, and progress.md in the project directory, with templates for phase tracking, research storage, and session logging. - Hook-based context injection: UserPromptSubmit, PreToolUse, PostToolUse, Stop, and PreCompact hooks re-inject plan state automatically, with SHA-256 attestation to block tampered plan content. - Parallel and autonomous modes: Supports isolated plans under .planning/ for concurrent tasks, plus autonomous and gated modes with ledger summaries and a completion gate for long-running agent loops. - Use Case: An agent running a multi-hour refactor is interrupted by context compaction; on the next turn it re-reads task_plan.md and progress.md, verifies the attested plan hash, and continues from the exact in-progress phase. ## Quick Start Ask the agent to plan out your multi-step project using planning files, for example: "Create a task plan with phases for refactoring the authentication module and track progress in planning files."
