
Planning with Files (Traditional Chinese)
Keeps multi-step AI work on disk as plan, findings and progress files that survive long sessions
- Repository
- OthmanAdi/planning-with-files
- Stars
- 27,822
- Licence
- MIT
- Files
- 15 files
What it does
When installed, the agent keeps a persistent set of Markdown files (task_plan.md, findings.md, progress.md) in your project as a working memory for complex, multi-step tasks, and re-reads them before major decisions so the goal stays in view. Lifecycle hooks inject the selected plan into the agent's context and can warn if a plan was altered after being locked with a hash attestation. This Traditional Chinese variant provides the same workflow with Chinese-language templates and trigger phrases.
Great for
- Developers or writers running long research or coding tasks that exceed about five tool calls
- People who want a visible, editable plan and progress log they can review and resume later
- Teams running several parallel tasks who need each to pin its own plan folder
Skip it if
- Short one-off questions, since the planning overhead adds little value
- Users who want the agent to automatically browse past session history: the skill only reads past sessions when explicitly asked, and replay output must be treated as untrusted data
- Sensitive projects where extra files and hook scripts running on every tool call are unwelcome without review
Try asking
- “幫我規劃這個多步驟的專案遷移”
- “建立任務計畫並拆解成階段”
- “記錄這次研究的發現到 findings.md”
- “恢復上次中斷的計畫並繼續進度”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 15 original files, the MIT licence and an attribution note.
planning-with-files-zht.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip planning-with-files-zht.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip planning-with-files-zht.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Pin a plan with PLAN_ID when running parallel tasks, or use separate git worktrees, so tasks do not overwrite each other's plans.
- Run attest-plan.sh only after finalising a plan if you want tamper warnings; clear the attestation to edit it again.
- Review the hook commands in SKILL.md before installing, since they run shell scripts on every prompt, tool call and stop event.
What's inside
- plan-inventory.md
- scripts/attest-plan.sh
- scripts/check-complete.sh
- scripts/gate-stop.sh
- scripts/init-session.sh
- scripts/ledger-append.sh
- scripts/ledger-summary.sh
- scripts/phase-status.sh
- scripts/plan-doctor.sh
- scripts/resolve-plan-dir.sh
- scripts/session-catchup.py
- scripts/set-active-plan.sh
- scripts/skill-hook.sh
- SKILL.md
- templates/findings.md
Redistributed unmodified from OthmanAdi/planning-with-files under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.