planning-with-files

Maintains task plans, findings, and progress logs in workspace markdown files.

284|54|Updated Jun 23, 2026
One-click install
npx skills add https://github.com/nomifun/nomifun-desktop --skill planning-with-files-nomifun
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: planning-with-files
Source: https://github.com/nomifun/nomifun-desktop/tree/main/crates/backend/nomifun-app/assets/builtin-skills/planning-with-files
Command: npx skills add https://github.com/nomifun/nomifun-desktop --skill planning-with-files-nomifun

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long-running or research-heavy tasks often lose important decisions, evidence, and progress when context windows fill up or sessions end. This Skill persists plans, verified findings, and chronological progress logs as workspace files so work can be resumed, audited, or handed off without losing state. ## Core Features & Use Cases - Durable Task Plans: Creates and maintains task_plan.md with objectives, scope, milestones, decisions, and completion criteria, keeping at most one milestone in progress. - Evidence and Progress Tracking: Records verified facts in findings.md and chronological work logs in progress.md, separating facts from hypotheses and preserving an audit trail. - Handoff and Resumption: Detects existing planning files before writing, supports multi-writer ownership rules, and leaves concise final handoffs with validation results. - Use Case: During a multi-session codebase migration, an agent records each verified finding and completed milestone in workspace files, so a new session or a different agent can resume exactly where the work stopped. ## Quick Start Use the planning-with-files skill to organize this complex task with persistent planning files in the workspace.

Frequently Asked Questions about planning-with-files

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I keep track of progress on a long-running AI task?

Create persistent planning files in the workspace: a task_plan.md for goals and milestones, a findings.md for verified evidence, and a progress.md for a chronological work log. Update them after meaningful changes so any session can resume the work.

What files should an AI agent write for task planning?

Three markdown files cover most needs: task_plan.md holds objectives, scope, milestones, and decisions; findings.md stores verified facts and open questions; progress.md logs actions, validation results, and handoff notes chronologically.

When should I not use file-based task planning?

Skip it for short tasks that fit comfortably in the active conversation, such as simple edits or brief explanations. Persistent planning files add overhead that only pays off for multi-phase, cross-session, or evidence-heavy work.

Can multiple agents share the same planning files?

Yes, but only one writer should update the planning files at a time. Hand off ownership explicitly before another agent writes, and use task-specific filenames or subdirectories if existing files belong to a different task.

What should never go into planning or progress files?

Never store secrets, access tokens, private keys, or unnecessary personal data. Sanitize command output before recording it, keep only the useful failure signature, and summarize large raw output instead of pasting full logs.