Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Create Plan · davila7 | DecimalAI

davila7/create-plan

SkillScore55Maintained

Create a concise plan. Use when a user explicitly asks for a plan related to a coding task.

productivityby davila7

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Project Management pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/davila7-create-plan/SKILL.md

decimalai skills pull davila7-create-plan --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/davila7-create-plan/SKILL.md https://app.decimal.ai/s/davila7-create-plan/SKILL.md

Raw file: /s/davila7-create-plan/SKILL.md ↗ · machine-readable summary at /s/davila7-create-plan.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 47d ago

—-32% tokens0% turns55% of whole cases pass with skill (12/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-09	✗→✓	▲ Improved	-17%	0%
case-11	✗→✓	▲ Improved	-45%	0%
case-12	✗→✓	▲ Improved	-38%	0%
case-13	✗→✓	▲ Improved	-27%	0%
case-14	✗→✓	▲ Improved	-26%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

davila7 / claude-code-templates

cli-tool/components/skills/development/create-plan/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

Skill contents

SKILL.md

Create Plan

Goal

Turn a user prompt into a single, actionable plan delivered in the final assistant message.

Minimal workflow

Throughout the entire workflow, operate in read-only mode. Do not write or update files.

Scan context quickly

Read README.md and any obvious docs (docs/, CONTRIBUTING.md, ARCHITECTURE.md).
Skim relevant files (the ones most likely touched).
Identify constraints (language, frameworks, CI/test commands, deployment shape).

Ask follow-ups only if blocking

Ask at most 1–2 questions.
Only ask if you cannot responsibly plan without the answer; prefer multiple-choice.
If unsure but not blocked, make a reasonable assumption and proceed.

Create a plan using the template below

Start with 1 short paragraph describing the intent and approach.
Clearly call out what is in scope and what is not in scope in short.
Then provide a small checklist of action items (default 6–10 items).
Each checklist item should be a concrete action and, when helpful, mention files/commands.
Make items atomic and ordered: discovery → changes → tests → rollout.
Verb-first: "Add…", "Refactor…", "Verify…", "Ship…".
Include at least one item for tests/validation and one for edge cases/risk when applicable.
If there are unknowns, include a tiny Open questions section (max 3).

Do not preface the plan with meta explanations; output only the plan as per template

Plan template (follow exactly)

markdown
# Plan

<1–3 sentences: what we're doing, why, and the high-level approach.>

## Scope
- In:
- Out:

## Action items
[ ] <Step 1>
[ ] <Step 2>
[ ] <Step 3>
[ ] <Step 4>
[ ] <Step 5>
[ ] <Step 6>

## Open questions
- <Question 1>
- <Question 2>
- <Question 3>

Checklist item guidance

Good checklist items:

Point to likely files/modules: src/..., app/..., services/...
Name concrete validation: "Run npm test", "Add unit tests for X"
Include safe rollout when relevant: feature flag, migration plan, rollback note

Avoid:

Vague steps ("handle backend", "do auth")
Too many micro-steps
Writing code snippets (keep the plan implementation-agnostic)

Illustrative Want to see what davila7/create-plan would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/12/2026, 14:34 UTC · 22 cases

Measured on 15 comparable cases — below the 22 needed to publish a lift, so this result is shown here but not on the registry card or leaderboard.

Lift

+5%

55% vs 50%

Agent turns

0%

1.0 vs 1.0

Tokens

-32%

1,974.4 vs 2,891.67

fail → pass · 7pass → pass · 5fail → fail · 4pass → fail · 6

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We need to integrate Google OAuth2 login into our web application backend. Before making any file modifications, please explore our repository context, documentation, and existing auth logic, then formulate a comprehensive implementation plan. The plan should outline the overall approach, explicitly specify what is in and out of scope, list sequential actionable tasks with checkboxes for development, testing, and deployment, and include any remaining unresolved questions.	fail→fail	4,228	3,104	-27%	1	1	0%	197	846	+329%	0	0	—
case-02 I want to migrate our current REST API endpoints to use GraphQL schema and resolvers. Please scan through the codebase and documentation to understand our setup, then lay out a structured rollout plan. Provide a concise summary of the approach, define the exact scope limits, give me an ordered checklist of task items covering investigation through testing and rollout, and raise any blocking questions if necessary.	fail→fail	15,171	2,808	-81%	1	1	0%	2,418	747	-69%	0	0	—
case-03 Our team is planning to add Redis-based rate limiting to the payment processing microservice. Before we start coding, analyze the repository structure and give us a step-by-step roadmap for this feature. Please present this as a clean plan that summarizes the core approach, highlights in-scope versus out-of-scope boundaries, provides a checklist of concrete action steps, and notes any open questions.	fail→fail	18,728	2,830	-85%	1	1	0%	2,859	771	-73%	0	0	—
case-04 Our JWT verification logic in src/auth/verify.ts has a memory leak. Refactor the code directly and return the updated file contents so we can commit it right away.	pass→fail	10,296	11,724	+14%	1	1	0%	1,771	2,073	+17%	0	0	—
case-05 Write a complete executable Python script using SQLAlchemy to migrate user profile records from MySQL to PostgreSQL.	pass→fail	39,863	9,172	-77%	1	1	0%	3,681	1,942	-47%	0	0	—
case-06 Fix the syntax error and signature validation logic in services/billing.py and output the corrected file content.	pass→fail	14,697	15,042	+2%	1	1	0%	2,890	992	-66%	0	0	—
case-07 We want to update our CI pipeline to support multi-architecture Docker builds. Analyze our workflow setup and generate a planning document. Ensure the task list is clear and action-oriented.	fail→fail	18,555	4,411	-76%	1	1	0%	2,921	765	-74%	0	0	—
case-08 We are adding database connection pooling to our Express backend. Please review the codebase and draft an implementation plan covering discovery through deployment.	pass→fail	18,916	3,057	-84%	1	1	0%	3,356	774	-77%	0	0	—
case-22 Draft an actionable plan to migrate our test runner from Jest to Vitest.	pass→fail	16,935	5,211	-69%	1	1	0%	3,137	805	-74%	0	0	—
case-09 We need to integrate Stripe webhook handling for subscription renewals. There are multiple unknown parameters regarding edge cases. Outline a technical plan and document unresolved items.	fail→pass	15,870	10,430	-34%	1	1	0%	2,517	2,095	-17%	0	0	—
case-10 Draft a plan for converting our frontend state management from Redux to Zustand. You might be tempted to show sample Zustand store definitions; avoid inline code blocks in the plan itself.	pass→fail	15,725	8,434	-46%	1	1	0%	2,477	1,810	-27%	0	0	—
case-11 Formulate an implementation plan for adding full-text search using Elasticsearch to our product catalog service.	fail→pass	19,713	7,576	-62%	1	1	0%	3,322	1,815	-45%	0	0	—
case-12 Create a roadmap for upgrading React 17 to React 18 in our legacy dashboard repository. Clearly define scope boundaries.	fail→pass	18,866	9,297	-51%	1	1	0%	3,239	2,012	-38%	0	0	—
case-13 Provide a step-by-step technical plan for adding Sentry error logging to our Node.js microservices. Do not write introductory prose like 'Sure, here is your plan'.	fail→pass	16,842	9,342	-45%	1	1	0%	2,858	2,075	-27%	0	0	—
case-14 Design a rollout plan for introducing gRPC communication between our user and order services.	fail→pass	17,400	10,686	-39%	1	1	0%	2,674	1,989	-26%	0	0	—
case-15 We are setting up automated database backups to AWS S3. Create an actionable step-by-step checklist plan for this infrastructure task.	fail→pass	15,419	8,106	-47%	1	1	0%	2,625	1,777	-32%	0	0	—
case-16 Create a technical plan for migrating user passwords from SHA-256 to Argon2id hashing upon next login.	pass→pass	24,201	8,670	-64%	1	1	0%	4,083	1,888	-54%	0	0	—
case-17 Plan the addition of a health check endpoint to our FastAPI service.	pass→pass	14,333	12,147	-15%	1	1	0%	2,565	2,391	-7%	0	0	—
case-18 Develop an execution plan for replacing Axios with native fetch across our web client modules.	pass→pass	22,317	8,130	-64%	1	1	0%	3,799	1,770	-53%	0	0	—
case-19 Formulate a strategy to clean up unused npm dependencies across our monorepo without modifying package.json right now.	pass→pass	39,581	19,378	-51%	1	1	0%	2,569	2,304	-10%	0	0	—
case-20 Create an implementation plan for adding dark mode theme support to our Tailwind CSS web app.	fail→pass	15,558	7,706	-50%	1	1	0%	2,696	1,787	-34%	0	0	—
case-21 Outline a plan for swapping our primary SMS provider from Twilio to MessageBird.	pass→pass	15,779	8,658	-45%	1	1	0%	2,499	1,888	-24%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 15 counted toward the lift figure. The other 7 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +5 percentage points is the difference between those two pass rates over the 15 comparable cases. 8 cases got worse with the skill loaded, and they are included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 50%
Pass rate with the skill: 54.6%
Cases attempted: 22
Cases counted toward lift: 15
Cases that regressed: 8
Skill version measured: v1
Run completed: 8/12/2026, 14:34 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-09

Prompt

We need to integrate Stripe webhook handling for subscription renewals. There are multiple unknown parameters regarding edge cases. Outline a technical plan and document unresolved items.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run55

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-34% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 46 days ago

Other measured skills in the registry, with their headline benchmark lift.