agentleFS
knwoopknwoop/clear-plan-skills/skills/update-plan/SKILL.md

update-plan

Skill9 starsChanged 29 days agoMITView on GitHub

Updates an existing implementation plan based on new information discovered during development. Merges the user's current plan with new findings (edge cases, API changes, scope changes, discovered dependencies) and outputs an updated plan, then copies it to the clipboard. Use this skill when the user says "update the plan", "revise the plan", "plan changed", "found a new edge case", "the approach needs to change", "replan", "adjust TODO", or pastes an existing plan and describes what changed. Also trigger when the user reports implementation surprises, blockers, or scope adjustments that affect the plan.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. About 0 are common, 10 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Before doing anything else, enter Plan Mode (use the EnterPlanMode tool if you are not already in it).

    line 23
  2. Very specific:

    If $ARGUMENTS is provided, use it as the description of what changed.

    line 34
  3. Distinctive:

    If empty, use the conversation context to understand the changes.

    line 35
  4. Distinctive:

    If neither provides enough detail, ask the user what changed.

    line 36
  5. Distinctive:

    If the current plan is not provided in the conversation or arguments, ask the user to choose:

    line 43
  6. Distinctive:

    Before the full plan, provide a brief summary of what changed:

    line 83
  7. Distinctive:

    If the script exits non-zero (no clipboard tool available), output the plan in a single fenced code block so the user can copy it manually.

    line 107
  8. Very specific:

    Do not alter completed [x] items unless the user explicitly asks

    line 115
  9. Distinctive:

    When splitting a TODO, preserve the original intent and add sub-tasks.

    line 116
  10. Distinctive:

    If the change invalidates a decision table row, update it — don't just add rows.

    line 118
  11. Distinctive:

    Keep conditions as yes/no questions and re-check that every combination is covered

    line 119
  12. Distinctive:

    If the user provides only partial plan text, ask for the missing sections before generating the update (or note assumptions clearly)

    line 120
  13. Distinctive:

    Keep the same format as the original plan — consistency matters for the workflow

    line 122

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Update Plan

Revise an existing plan based on new information and copy the updated version to the clipboard.

IMPORTANT: Enter Plan Mode first

Before doing anything else, enter Plan Mode (use the EnterPlanMode tool if you are not already in it). Do not write or modify any code. Plan Mode allows you to explore the codebase read-only and revise the plan without making changes.

Process

Step 1: Receive the Current Plan and Changes

What changed: $ARGUMENTS

If $ARGUMENTS is provided, use it as the description of what changed. If empty, use the conversation context to understand the changes. If neither provides enough detail, ask the user what changed.

The user provides:

  1. The current plan — pasted from their notes (full or partial)
  2. What changed — described in natural language

If the current plan is not provided in the conversation or arguments, ask the user to choose:

  • Paste the current plan
  • Infer from conversation context

Common types of changes:

Change TypeExample
New edge case"The API returns 404 when the resource is soft-deleted, we need to handle this"
Scope adjustment"We decided not to implement restore for v1"
Approach change"The ORM doesn't support bulk upsert, need to use raw SQL"
New dependency"This requires a migration to add the deleted_at column first"
Task splitting"The repository TODO is too large, needs to be split" (split by feature within the layer)
PR splitting"PR 2 is too big to review" (split by behavior into PR 2 and PR 3, renumber later PRs)
Completed items"I finished the domain and repository tasks"
Criteria change"Restore should also work on items deleted by an admin"

Step 2: Generate the Updated Plan

Apply the changes to the plan while following these rules:

  1. Preserve completed items — Keep [x] items as-is for history
  2. Mark removed items — Use strikethrough with a reason, don't silently delete
  3. Add new items at the correct dependency position — not just appended at the end
  4. Update What — if the goal or scope fundamentally changed, revise it
  5. Update the decision tables — if edge cases changed, update the table in the affected TODO's "Done when" (or the shared table, if several TODOs follow it)
  6. Update references — add new files, PRs, or documents discovered during work
  7. Update requirements — if the specifications or constraints changed
  8. Keep "Done when" and "Verify" in sync — when a TODO's approach or scope changes, update its "Done when" and "Verify". When a shared decision table row is added, removed, or renumbered, update every open TODO that cites it. New TODOs need both fields

Output the full updated plan following the same five-section format. The format specification is shared with create-plan; read it at ${CLAUDE_SKILL_DIR}/../create-plan/references/plan-format.md.

Step 3: Show the Diff Summary

Before the full plan, provide a brief summary of what changed:

### Plan Changes

- Added: <new TODO items or sections>
- Modified: <changed approach or scope>
- Removed: <items no longer needed, with reason>
- Reordered: <dependency changes>

This helps the user quickly verify the update matches their intent.

Step 4: Copy to Clipboard

Copy the updated plan (without the diff summary) to the clipboard with the clipboard script bundled in create-plan:

"${CLAUDE_SKILL_DIR}/../create-plan/scripts/copy-to-clipboard.sh" <<'PLAN_EOF'
<the updated plan>
PLAN_EOF

If the script exits non-zero (no clipboard tool available), output the plan in a single fenced code block so the user can copy it manually.

Inform the user the updated plan is on their clipboard so they can replace the old version in their notes.

Guidelines

  • Do not alter completed [x] items unless the user explicitly asks
  • When splitting a TODO, preserve the original intent and add sub-tasks; divide the original "Done when" between them so no criterion is lost
  • If the change invalidates a decision table row, update it — don't just add rows. Keep conditions as yes/no questions and re-check that every combination is covered
  • If the user provides only partial plan text, ask for the missing sections before generating the update (or note assumptions clearly)
  • Keep the same format as the original plan — consistency matters for the workflow

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.