home / skills / eyadsibai / ltk / planning-with-files

planning-with-files skill

/plugins/ltk-core/skills/planning-with-files

This skill helps you kick off complex tasks with file-based planning, ensuring clear goals, phased progress, and persistent findings.

npx playbooks add skill eyadsibai/ltk --skill planning-with-files

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
5.0 KB
---
name: planning-with-files
description: This skill should be used when starting complex multi-step tasks, research projects, or any task requiring >5 tool calls. Implements Manus-style file-based planning with task_plan.md, findings.md, and progress.md.
version: 1.0.0
---

# Planning with Files

Work like Manus: Use persistent markdown files as your "working memory on disk."

## The Core Pattern

```
Context Window = RAM (volatile, limited)
Filesystem = Disk (persistent, unlimited)

Anything important gets written to disk.
```

## Quick Start

Before ANY complex task:

1. **Create `task_plan.md`** - Use the template below
2. **Create `findings.md`** - For research and discoveries
3. **Create `progress.md`** - For session logging
4. **Re-read plan before decisions** - Refreshes goals in attention window
5. **Update after each phase** - Mark complete, log errors

## File Purposes

| File | Purpose | When to Update |
|------|---------|----------------|
| `task_plan.md` | Phases, progress, decisions | After each phase |
| `findings.md` | Research, discoveries | After ANY discovery |
| `progress.md` | Session log, test results | Throughout session |

## Critical Rules

### 1. Create Plan First

Never start a complex task without `task_plan.md`. Non-negotiable.

### 2. The 2-Action Rule

> "After every 2 view/browser/search operations, IMMEDIATELY save key findings to text files."

This prevents visual/multimodal information from being lost.

### 3. Read Before Decide

Before major decisions, read the plan file. This keeps goals in your attention window.

### 4. Update After Act

After completing any phase:

- Mark phase status: `in_progress` -> `complete`
- Log any errors encountered
- Note files created/modified

### 5. Log ALL Errors

Every error goes in the plan file. This builds knowledge and prevents repetition.

```markdown
## Errors Encountered
| Error | Attempt | Resolution |
|-------|---------|------------|
| FileNotFoundError | 1 | Created default config |
| API timeout | 2 | Added retry logic |
```

### 6. Never Repeat Failures

```
if action_failed:
    next_action != same_action
```

Track what you tried. Mutate the approach.

## The 3-Strike Error Protocol

```
ATTEMPT 1: Diagnose & Fix
  -> Read error carefully
  -> Identify root cause
  -> Apply targeted fix

ATTEMPT 2: Alternative Approach
  -> Same error? Try different method
  -> Different tool? Different library?
  -> NEVER repeat exact same failing action

ATTEMPT 3: Broader Rethink
  -> Question assumptions
  -> Search for solutions
  -> Consider updating the plan

AFTER 3 FAILURES: Escalate to User
  -> Explain what you tried
  -> Share the specific error
  -> Ask for guidance
```

## Read vs Write Decision Matrix

| Situation | Action | Reason |
|-----------|--------|--------|
| Just wrote a file | DON'T read | Content still in context |
| Viewed image/PDF | Write findings NOW | Multimodal -> text before lost |
| Browser returned data | Write to file | Screenshots don't persist |
| Starting new phase | Read plan/findings | Re-orient if context stale |
| Error occurred | Read relevant file | Need current state to fix |
| Resuming after gap | Read all planning files | Recover state |

## The 5-Question Reboot Test

If you can answer these, your context management is solid:

| Question | Answer Source |
|----------|---------------|
| Where am I? | Current phase in task_plan.md |
| Where am I going? | Remaining phases |
| What's the goal? | Goal statement in plan |
| What have I learned? | findings.md |
| What have I done? | progress.md |

## Templates

### task_plan.md Template

```markdown
# Task Plan: [Task Name]

## Goal
[Clear statement of what we're trying to accomplish]

## Phases

### Phase 1: [Name]
- Status: pending | in_progress | complete
- Tasks:
  - [ ] Task 1
  - [ ] Task 2

### Phase 2: [Name]
- Status: pending
- Tasks:
  - [ ] Task 1

## Errors Encountered
| Error | Attempt | Resolution |
|-------|---------|------------|

## Decisions Made
| Decision | Rationale | Date |
|----------|-----------|------|

## Files Modified
- [file1.py] - [what changed]
```

### findings.md Template

```markdown
# Research Findings

## [Topic 1]
- Key insight 1
- Key insight 2
- Source: [where found]

## [Topic 2]
- Finding 1
- Finding 2
```

### progress.md Template

```markdown
# Session Progress

## [Date/Time]
- Started: [task]
- Completed: [what]
- Blocked on: [if any]
- Next: [what's next]
```

## When to Use This Pattern

**Use for:**

- Multi-step tasks (3+ steps)
- Research tasks
- Building/creating projects
- Tasks spanning many tool calls
- Anything requiring organization

**Skip for:**

- Simple questions
- Single-file edits
- Quick lookups

## Anti-Patterns

| Don't | Do Instead |
|-------|------------|
| Use TodoWrite for persistence | Create task_plan.md file |
| State goals once and forget | Re-read plan before decisions |
| Hide errors and retry silently | Log errors to plan file |
| Stuff everything in context | Store large content in files |
| Start executing immediately | Create plan file FIRST |
| Repeat failed actions | Track attempts, mutate approach |

Overview

This skill implements a Manus-style file-based planning workflow for complex, multi-step tasks. It creates and maintains three persistent markdown files—task_plan.md, findings.md, and progress.md—to keep important context on disk and out of volatile memory. Use it to structure large tasks, preserve discoveries, and enforce disciplined error handling.

How this skill works

Before starting any complex task, the skill generates task_plan.md, findings.md, and progress.md using provided templates. During work it enforces the 2-action rule: after two view/search actions, key findings are written to files. It updates phase statuses, logs errors and attempts, and re-reads plans before major decisions to keep goals in the active attention window. The skill also implements the 3-strike error protocol to escalate after repeated failures.

When to use it

  • Starting complex tasks that require five or more tool calls
  • Research projects or literature reviews needing persistent notes
  • Multi-phase builds, experiments, or integrations
  • Work that spans multiple sessions or long time gaps
  • Situations where multimodal results (images/PDFs) must be captured to text

Best practices

  • Always create task_plan.md before executing substantive steps
  • After every two view/browser/search actions, immediately write key findings to findings.md
  • Re-read task_plan.md before making any major decision or changing approach
  • Log every error and attempted fix in the plan file; follow the 3-strike protocol before escalating
  • Mark phase statuses (pending, in_progress, complete) and list files modified after each phase

Example use cases

  • Orchestrating a multi-component software feature with design, implementation, and test phases
  • Conducting literature review and consolidating sources and insights across many web searches
  • Running an experiment suite where each run’s results and errors must be recorded persistently
  • Performing a long-running data migration that requires repeated tool calls and retry logic
  • Collaborative projects where teammates need a single source of truth on progress and decisions

FAQ

What if I forget to write findings after two views?

Treat it as an error: add the missing findings immediately to findings.md, note the lapse in progress.md, and repeat the 2-action discipline going forward.

When should I escalate to the user?

After three failed attempts at the same action (3-strike protocol). Summarize what you tried, the errors, and proposed next steps in the plan before asking for guidance.