ds-plan-loop — Claude Code skill: planning and setup
The ds-plan-loop Claude Code skill builds a plan with a check for each step, reviews it, then carries out the task one step at a time. It activates for multi-file work, unclear scope, or tasks that previously missed steps.
MIT 569 installs ★ 320 updated 2026-10-28
npx skills add aicodeking/deepseek-stack --skill ds-plan-loop -a claude-codeWhat ds-plan-loop does
The skill guides an agent to plan before editing code and review the plan in a separate pass. It first checks the goal and the files involved, then saves numbered steps and a verifiable check for each in PLAN.md or PROGRESS.md.
After the review, the agent resolves unchecked assumptions and works through the plan one step at a time, recording each check’s result. If a step fails its check twice in a row, the instructions say to switch to ds-unstick; ds-verify-loop is suggested for a failed check. Installation through npm uses npx, so Node.js is required.
What it helps with
- split multi-file work into steps with checks
- catch a missing step before editing code
- review a plan after testing its assumptions
- record check results as work proceeds
Example prompts for the agent
- Plan the changes across these files and complete one verified step at a time:
/ds-plan-loop. - Break down this unclear task, then review the plan before editing:
/ds-plan-loop. - We missed steps last time; make and review a plan before changing the code:
/ds-plan-loop.
How to install ds-plan-loop
1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-plan-loop -a claude-code2Manually: put the files in place
~/.claude/skills/ds-plan-loop/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/ds-plan-loop/
1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-plan-loop -a codex2Manually: put the files in place
.agents/skills/ds-plan-loop/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-plan-loop -a cursor2Manually: put the files in place
.cursor/skills/ds-plan-loop/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install aicodeking/deepseek-stack skills/ds-plan-loop2Manually: put the files in place
.github/skills/ds-plan-loop/SKILL.md1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-plan-loop -a gemini-cli2Platform CLI
gemini skills install https://github.com/aicodeking/deepseek-stack --path skills/ds-plan-loopManually: put the files in place
.devin/skills/ds-plan-loop/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-plan-loop -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about ds-plan-loop
How do I install ds-plan-loop in Claude Code?
Run npx skills add aicodeking/deepseek-stack --skill ds-plan-loop -a claude-code. The skill file is at ~/.claude/skills/ds-plan-loop/SKILL.md.
What does ds-plan-loop do?
It guides the agent to write a plan with a check for every step, review and revise it, then execute it step by step. Check results go in the plan file.
When does ds-plan-loop stop execution?
It directs the agent to stop if a step fails its check twice in succession and switch to ds-unstick. If the plan proves wrong, the remaining steps must be revised before work continues.
Source text
SKILL.md · MIT
--- name: ds-plan-loop description: "Write a numbered plan with an acceptance check for each step, critique it once in a fresh pass, revise it, then execute one step at a time. Use when a task touches more than one file, has unclear scope, or failed before because steps were missed. Invoke as /ds-plan-loop." --- # ds-plan-loop Plan before coding, check the plan once, then build the task one verified step at a time. ## Steps 1. Restate the goal in one sentence. List the files and commands you expect to touch. Read those files before you plan. Do not plan from memory. 2. Write a numbered plan. Each step gets one action and one acceptance check that a command or a direct observation can confirm, for example "`npm test -- auth` passes" or "`GET /health` returns 200". Keep each step small enough to verify in a single diff. 3. Write the plan to a file (PLAN.md, or the PROGRESS.md from ds-context-reset) so later steps and fresh sessions can read it. 4. Critique the plan in a fresh pass. Reread only the goal and the plan, as a reviewer who did not write it. If you can, run the critique in a subagent or new session. Answer three questions: - Which step is missing (setup, migration, config, tests, cleanup)? - Which assumption has not been checked against the code or docs? - Is there a simpler path that reaches the same goal with fewer steps? 5. Revise the plan. Check each risky assumption now by reading code or running a command. Do not carry it forward as a guess. 6. Execute step 1 only. Run its acceptance check and read the real output. Use ds-verify-loop when the check fails. 7. Mark the step done in the plan file, along with the evidence (command and result). Then move to the next step. 8. If a step shows that the plan is wrong, stop and revise the remaining steps before continuing. Do not improvise around it. ## Stop when - Every step is marked done and its check passed, and the original goal holds when you check it end to end. - You have done 2 critique rounds. A third round almost never finds anything new, so execute. - A step fails its check twice in a row. Switch to ds-unstick. ## Rules - No step without a check. "Looks right" is not a check. - Do one step per diff. Do not batch steps to save time. - Do not add steps that the goal does not need. - Keep the plan short: about 3-10 steps. Split bigger work into separate plans.
© aicodeking · MIT · aicodeking/deepseek-stack