file-context-planning — file-context-planning install file-context-planning, hookcode, community, file-context-planning install, ide skills, Codex-compatible skills, CLI automation tools, code review and automation platforms, Claude Code, Cursor, Windsurf

v4.0.0
GitHub

About this Skill

Perfect for Development Agents needing advanced code review and automation through file context planning. file-context-planning is a Codex-compatible skill that uses persistent markdown files as working memory on disk for planning and automation purposes.

Features

Triggers CLI coding assistants through conversations and Webhooks
Supports independent deployment for flexible automation
Provides a visual console for real-time task execution monitoring
Utilizes persistent markdown files as working memory on disk
Compatible with Codex for instruction-only automation
Refactors planning files for traceability under docs/en/developer/plans/<hash>

# Core Topics

hookvibe hookvibe
[3]
[1]
Updated: 2/27/2026

Agent Capability Analysis

The file-context-planning skill by hookvibe is an open-source community AI agent skill for Claude Code and other IDE workflows, helping agents execute tasks with better context, repeatability, and domain-specific guidance. Optimized for file-context-planning install, Codex-compatible skills, CLI automation tools.

Ideal Agent Persona

Perfect for Development Agents needing advanced code review and automation through file context planning.

Core Value

Empowers agents to perform intelligent code reviews and automate tasks through conversations and Webhooks, utilizing visual consoles for real-time task execution monitoring and persistent markdown files as working memory on disk.

Capabilities Granted for file-context-planning

Automating code reviews with real-time task monitoring
Generating planning files for traceability and version control
Debugging code execution with visual console outputs

! Prerequisites & Limits

  • Requires filesystem access for persistent markdown files
  • Limited to Codex-compatible skills and instruction-only workflows
Labs Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

file-context-planning

Install file-context-planning, an AI agent skill for AI agent workflows and automation. Works with Claude Code, Cursor, and Windsurf with one-command setup.

SKILL.md
Readonly

Planning with Files (Codex)

{/* Align this Claude skill's path examples with the .claude directory layout. docs/en/developer/plans/claude-skills-audit-20260227/task_plan.md claude-skills-audit-20260227 /} Work like Manus: Use persistent markdown files as your "working memory on disk." {/ Refactor planning files to live under docs/en/developer/plans/<hash> for traceability. sddsa89612jk4hbwas678 */}

What This Skill Does (Codex-Compatible)

Codex skills are instruction-only. There are no automatic hooks (e.g., "PreToolUse", "Stop").

To get the benefits, you must manually:

  • Create a session folder under docs/en/developer/plans/<session-hash>/
  • Re-read task_plan.md before major decisions
  • Update task_plan.md, findings.md, and progress.md throughout the work
  • Add the same <session-hash> into inline code comments for traceability

Operational Checklist (Replace Claude Hooks)

When using this skill on a complex task:

  1. Create a new session hash (example: sddsa89612jk4hbwas678).
  2. If docs/en/developer/plans/<session-hash>/ does not exist, initialize it (script or template copy).
  3. Fill task_plan.md (goal, phases, key questions) before implementing anything.
  4. Before major decisions or phase changes, re-read task_plan.md (refreshes goals into the attention window).
  5. After every 2 information-gathering actions, append key discoveries into findings.md (2-Action Rule).
  6. After completing a phase, return to the session folder and update:
    • phase status in task_plan.md
    • session log + tests in progress.md
  7. When the task is done, update docs/en/change-log/0.0.0.md with the hash + one-line summary + relative link to the plan.

Important: Where Files Go (Repo Layout)

When using this skill:

  • Skill assets (templates/scripts/reference) live in: .claude/skills/file-context-planning/
    • Templates: .claude/skills/file-context-planning/templates/
    • Scripts: .claude/skills/file-context-planning/scripts/
  • Your planning files live in a hash folder under: docs/en/developer/plans/<session-hash>/
LocationWhat Goes There
Skill directory (.claude/skills/file-context-planning/)Templates, scripts, reference docs
docs/en/developer/plans/<session-hash>/task_plan.md, findings.md, progress.md

This ensures your planning files are:

  • durable (versionable docs)
  • easy to link to (relative links)
  • traceable from code via the session hash

Quick Start

Before ANY complex task:

Option A: Initialize via Script (Fastest)

Creates docs/en/developer/plans/<session-hash>/ and copies templates there if missing. {/* Document docs navigation sync behavior (Mintlify vs Docusaurus). docs/en/developer/plans/dsim8xybp9oa18nz1gfq/task_plan.md dsim8xybp9oa18nz1gfq */} Also syncs docs/docs.json when present (Mintlify) so the new session files are discoverable; if docs/docs.json is absent (Docusaurus), the sync step is skipped.

bash
1bash .claude/skills/file-context-planning/scripts/init-session.sh "<session-hash>" "<session-title>"

If you need a one-off backfill or want to rebuild the navigation index deterministically:

bash
1bash .claude/skills/file-context-planning/scripts/sync-docs-json-plans.sh

Option B: Copy Templates (Most Detailed)

bash
1mkdir -p docs/en/developer/plans/<session-hash> 2cp .claude/skills/file-context-planning/templates/task_plan.md docs/en/developer/plans/<session-hash>/task_plan.md 3cp .claude/skills/file-context-planning/templates/findings.md docs/en/developer/plans/<session-hash>/findings.md 4cp .claude/skills/file-context-planning/templates/progress.md docs/en/developer/plans/<session-hash>/progress.md

Then:

  1. Fill task_plan.md first — Goal, phases, key questions
  2. Re-read plan before decisions — Refreshes goals in the attention window
  3. Update after each phase — Mark status complete, log errors
  4. Log discoveries continuouslyfindings.md for knowledge, progress.md for session/test logs

Optional completion check:

bash
1bash .claude/skills/file-context-planning/scripts/check-complete.sh <session-hash>

Note: All three planning files should be created in docs/en/developer/plans/<session-hash>/, not in your project root and not in the skill folder.

The Core Pattern

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

→ Anything important gets written to disk.

File Purposes

FilePurposeWhen to Update
task_plan.mdPhases, progress, decisionsAfter each phase
findings.mdResearch, discoveriesAfter ANY discovery
progress.mdSession log, test resultsThroughout session

Traceability: Hash in Inline Comments

To link code changes back to the plan folder, include the session hash in every changed area via an inline comment:

{/* Align traceability format with repo-root AGENTS.md (include plan path + hash). docs/en/developer/plans/docsworkflowapi20260121/task_plan.md docsworkflowapi20260121 */} Format: <one sentence in English> <relative-plan-path> <session-hash>

Plan path: docs/en/developer/plans/<session-hash>/task_plan.md

Examples:

  • JS/TS/Go: // Add input validation for webhook payload. docs/en/developer/plans/sddsa89612jk4hbwas678/task_plan.md sddsa89612jk4hbwas678
  • Python/Shell/YAML: # Document retry backoff behavior. docs/en/developer/plans/sddsa89612jk4hbwas678/task_plan.md sddsa89612jk4hbwas678
  • SQL: -- Prevent duplicate inserts via unique key. docs/en/developer/plans/sddsa89612jk4hbwas678/task_plan.md sddsa89612jk4hbwas678
  • Markdown: {/* Explain why this doc section changed. docs/en/developer/plans/sddsa89612jk4hbwas678/task_plan.md sddsa89612jk4hbwas678 */}

This creates a stable backlink: comment → hash → docs/en/developer/plans/<hash>/.

Release Note Update (Required)

When the task is completed, update docs/en/change-log/0.0.0.md with:

  • the session hash
  • a one-line summary
  • a relative link to the plan (recommend linking task_plan.md)

Example entry:

md
1- sddsa89612jk4hbwas678: Refactor file-context-planning to store plans in hash folders. ([plan](../developer/plans/sddsa89612jk4hbwas678/task_plan.md))

{/* Keep changelog entries clean and single-line (no extra HTML comment lines). l290bb7v758opd6uxu6r */}

Note: Do not add an extra {/* ... */} line above changelog bullets. The bullet itself (hash + plan link) is enough traceability.

{/* Prefer stdin to avoid shell expansion when summaries contain code-like characters. l290bb7v758opd6uxu6r */}

Tip: If your summary contains backticks or other shell-sensitive characters, pipe it via stdin: printf '%s' "<summary>" | bash .claude/skills/file-context-planning/scripts/append-changelog.sh "<hash>".

Critical Rules

1. Create Plan First

Never start a complex task without a session folder containing task_plan.md. Non-negotiable.

2. The 2-Action Rule

"After every 2 information-gathering actions, 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_progresscomplete
  • 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
1## Errors Encountered 2| Error | Attempt | Resolution | 3|-------|---------|------------| 4| FileNotFoundError | 1 | Created default config | 5| 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

SituationActionReason
Just wrote a fileDON'T readContent still in context
Viewed image/PDFWrite findings NOWMultimodal → text before lost
Browser returned dataWrite to fileScreenshots don't persist
Starting new phaseRead plan/findingsRe-orient if context stale
Error occurredRead relevant fileNeed current state to fix
Resuming after gapRead all planning filesRecover state

The 5-Question Reboot Test

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

QuestionAnswer 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

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

Templates

Copy these templates to start:

Scripts

Helper scripts for automation:

  • scripts/init-session.sh — Initialize all planning files
  • scripts/check-complete.sh — Verify all phases complete
  • scripts/append-changelog.sh — Append changelog entry (hash + plan link)

Advanced Topics

Anti-Patterns

Don'tDo Instead
Keep todos only in chat memoryPersist the plan in task_plan.md
State goals once and forgetRe-read plan before decisions
Hide errors and retry silentlyLog errors to plan file
Stuff everything in contextStore large content in files
Start executing immediatelyCreate plan file FIRST
Repeat failed actionsTrack attempts, mutate approach
Create planning files in repo rootCreate docs/en/developer/plans/<hash>/

FAQ & Installation Steps

These questions and steps mirror the structured data on this page for better search understanding.

? Frequently Asked Questions

What is file-context-planning?

Perfect for Development Agents needing advanced code review and automation through file context planning. file-context-planning is a Codex-compatible skill that uses persistent markdown files as working memory on disk for planning and automation purposes.

How do I install file-context-planning?

Run the command: npx killer-skills add hookvibe/hookcode/file-context-planning. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for file-context-planning?

Key use cases include: Automating code reviews with real-time task monitoring, Generating planning files for traceability and version control, Debugging code execution with visual console outputs.

Which IDEs are compatible with file-context-planning?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for file-context-planning?

Requires filesystem access for persistent markdown files. Limited to Codex-compatible skills and instruction-only workflows.

How To Install

  1. 1. Open your terminal

    Open the terminal or command line in your project directory.

  2. 2. Run the install command

    Run: npx killer-skills add hookvibe/hookcode/file-context-planning. The CLI will automatically detect your IDE or AI agent and configure the skill.

  3. 3. Start using the skill

    The skill is now active. Your AI agent can use file-context-planning immediately in the current project.

Related Skills

Looking for an alternative to file-context-planning or another community skill for your workflow? Explore these related open-source skills.

View All

widget-generator

Logo of f
f

f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.

149.6k
0
AI

flags

Logo of vercel
vercel

flags is a Next.js feature management skill that enables developers to efficiently add or modify framework feature flags, streamlining React application development.

138.4k
0
Browser

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI