Ready to supercharge your AI agents?

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

Plan Creation Rules · growthxai | DecimalAI

growthxai/output-plan-workflow

NewSkillScoreMaintained

Use when the user asks to create, build, generate, scaffold, or plan a new workflow. Orchestrates the full planning process including architecture, steps, prompts, evaluators, and testing strategy using specialized subagents.

agent-toolingby growthxai

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

Automatic

AI Engineer pack

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

pip install decimalai

2 · Get the skill — writes .claude/skills/growthxai-output-plan-workflow/SKILL.md

decimalai skills pull growthxai-output-plan-workflow --out .claude/skills/

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

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

Raw file: /s/growthxai-output-plan-workflow/SKILL.md ↗ · machine-readable summary at /s/growthxai-output-plan-workflow.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

23 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 15h ago

—+153% tokens0% turns65% of whole cases pass with skill (15/23)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-07	✗→✓	▲ Improved	58%	0%
case-08	✗→✓	▲ Improved	157%	0%
case-09	✗→✓	▲ Improved	341%	0%
case-10	✗→✓	▲ Improved	63%	0%
case-11	✗→✓	▲ Improved	86%	0%

Showing 5 of 23 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/17/2026

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

growthxai / output

coding_assistants/claude/plugins/outputai/skills/output-plan-workflow/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/17/2026Version v2

Skill contents

SKILL.md

Your task is to generate a comprehensive Output.ai workflow implementation plan in markdown format.

The plan will be displayed to the user who can then decide what to do with it.

Please respond with only the final version of the plan.

Use the todo tool to track your progress through the plan creation process.

Plan Creation Rules

Overview

Generate detailed specifications for implementation of a new workflow.

Output Path

All plan outputs go to: .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md

<process_flow>

Step 0: Arguments Analysis

Analyze the arguments the user provided:

Ensure they have provided:

workflow_description: The description of the workflow to be created
additional_instructions: Additional instructions for the workflow

If not, ask the user for the missing information. </substep>

<substep number="1" name="pre_flight_check"> EXECUTE: Claude Skill: output-meta-pre-flight </substep>

</step>

Step 1: Context Gathering

Take the time to gather all the context you need to create a comprehensive plan.

Read any given files or links
Find any related workflows in the project
Read the projects documentation files

</step>

Step 2: Requirements Clarification

Clarify scope boundaries and technical considerations by asking numbered questions as needed to ensure clear requirements before proceeding.

<clarification_areas> <scope>

in_scope: what is included
out_of_scope: what is excluded (optional)

</scope> <technical>

functionality specifics
UI/UX requirements
integration points

</technical> <llm_provider>

Ask which LLM provider the user wants to use (anthropic, openai, google-vertex, or amazon-bedrock)
Default to anthropic if the user has no preference
All prompt files in the workflow must use the same provider unless the user explicitly requests otherwise
Record the chosen provider so it flows through to prompt engineering (step 6) and implementation

</llm_provider> </clarification_areas>

<decision_tree> IF clarification_needed: ASK numbered_questions WAIT for_user_response ELSE: PROCEED schema_definition </decision_tree>

</step>

Step 3: Workflow Design

Design the workflow with clear single purpose steps and sound orchestration logic.

<thought_process>

Define the workflow name and description
What is a valid output schema for the workflow?
What is a valid input schema for the workflow?
What needs to happen to transform the input into the output?
What are the atomic steps that need to happen to transform the input into the output?
How do these steps relate to each other?
Are any of these steps conditional?
Are any of these steps already defined in the project?
How could these steps fail, and how should we handle them? (retry, backoff, etc.)

</thought_process>

<step_output> Output Draft Plan: to .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md </step_output>

</step>

Step 4: Step Design

Design the individual steps called by the workflow with clear boundaries.

<thought_process>

What is the name and description of each step?
What is the input schema for each step?
What is the output schema for each step?
What external services or APIs does each step use?
What error handling is needed for each step?
What retry policies should each step have?

</thought_process>

<step_output> Output Updated Plan: to .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md </step_output>

</step>

Step 4.5: Evaluator Design

Determine if the workflow requires quality assessment, validation, or content evaluation.

<decision_tree> IF workflow_outputs_need_quality_scoring: DESIGN evaluator functions IF workflow_has_llm_generated_content: CONSIDER content evaluation (factual accuracy, relevance, tone) IF workflow_requires_validation_with_confidence: DESIGN validation evaluators ELSE: SKIP evaluator design (note in plan: "No evaluators needed") </decision_tree>

<thought_process>

Does the workflow produce content that needs quality assessment?
Are there LLM-generated outputs that need evaluation?
Would the workflow benefit from confidence-scored validation?
What evaluation result types are appropriate (boolean/number/string)?
Should evaluators use simple logic or LLM-powered assessment?
Would offline eval testing with @outputai/evals be appropriate for dataset-driven verification?

</thought_process>

<step_output> Output Updated Plan: to .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md </step_output>

</step>

Step 5: Plan Review

Review the draft plan and make any necessary changes.

<thought_process>

Does the plan make sense?
Are all the steps clear and concise?
Are all the dependencies identified?
Does the workflow follow Output SDK conventions?
Are error handling patterns appropriate?
Is the input/output schema design correct?

</thought_process>

<decision_tree> IF changes_needed: UPDATE draft_plan ELSE: PROCEED to step 6 </decision_tree>

<step_output> Output Reviewed Plan: to .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md </step_output>

</step>

Step 6: Prompt Engineering

If any of the steps use an LLM, design the prompts for the steps.

<decision_tree> IF step_uses_llm: USE prompt_step_template ELSE: SKIP to step 7 </decision_tree>

<step_output> Output Updated Plan: to .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md </step_output>

</step>

Step 7: Testing Strategy

Design the testing strategy for the workflow.

<thought_process>

What unit tests do we need to write?
How can I run the workflow?
What cases do we need to validate?
What scenario files should be created?
Should we create eval datasets for offline testing with output workflow test?
What ground truth values are needed for eval datasets?

</thought_process>

<step_output> Output Updated Plan: to .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md </step_output>

</step>

Step 8: Generate Plan

Generate the complete plan in markdown format.

Note that every implementation should start with running the cli command npx output workflow generate --skeleton to create the workflow directory structure.

<file_template> <header> # Workflow Requirements Document

> Workflow: WORKFLOW_NAME] > Created: CURRENT_DATE] </header> <required_sections>

Overview
Spec Scope
Out of Scope
Workflow Design
Step Design
Evaluator Design (if applicable)
Prompt Design
Testing Strategy
Implementation Phases

</required_sections> </file_template>

<step_output> Output Final Plan: to .outputai/plans/YYYY_MM_DD_<workflow_name>_<task_name>/PLAN.md </step_output>

</step>

Step 9: Post-Flight Check

Verify the plan is complete and ready for implementation.

<substep number="0" name="post_flight_check"> EXECUTE: Claude Skill: output-meta-post-flight </substep>

Then instruct the user to:

Review the plan
Make any necessary changes
Implement the workflow by invoking the output-build-workflow skill, providing the plan file path, workflow name, and workflow directory

</step>

</process_flow>

---- START ----

Use the workflow description and any additional instructions the user provided.

Want to see if growthxai/output-plan-workflow would activate on your phrasing? Try it before you install — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/20/2026, 08:36 UTC · 23 cases

Measured on 17 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

+39%

65% vs 26%

Agent turns

1.0 vs 1.0

Tokens

+153%

2,969.82 vs 1,174.76

fail → pass · 13pass → pass · 2fail → fail · 4pass → fail · 4

Per-case results (23)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-06 There is a bug in our step code where timeout errors are not caught. Provide a TypeScript code snippet fixing the catch block in src/steps/parse.ts.	pass→fail	9,033	20,575	+128%	1	1	0%	1,436	5,786	+303%	0	0	—
case-07 In an Output workflow project, what exact relative directory path format and filename standard should be used when writing a generated markdown plan for a workflow named order-sync and task initial-spec?	fail→pass	10,064	3,382	-66%	1	1	0%	1,745	2,756	+58%	0	0	—
case-01 I want to create a new Output.ai workflow that automatically ingests customer support tickets, categorizes their urgency, and routes critical issues to Slack. Can you construct a full implementation spec and plan document in markdown format for this project, covering step structures, prompt definitions, and testing strategies?	fail→fail	25,352	36,029	+42%	1	1	0%	4,494	3,243	-28%	0	0	—
case-02 We need to design an Output.ai workflow that fetches nightly RSS feeds, summarizes key technical articles, and generates a formatted newsletter draft. Please generate a detailed workflow implementation plan in markdown detailing the input/output schemas, evaluation checks, and step orchestration so our engineering team can review it.	fail→fail	24,487	11,380	-54%	1	1	0%	4,147	2,890	-30%	0	0	—
case-03 Can you outline an implementation plan for an Output.ai financial processing workflow? The system should accept PDF invoices, parse billing fields using LLMs, verify calculated totals, and export structured JSON. Provide a structured markdown spec plan specifying individual steps, error retry handling, and scenario testing.	fail→fail	24,236	9,134	-62%	1	1	0%	3,913	2,770	-29%	0	0	—
case-04 We already have a fully approved spec for an Output SDK step that fetches data from a REST API. Write the TypeScript code for this step definition using the Output SDK.	pass→fail	17,879	13,466	-25%	1	1	0%	3,626	3,439	-5%	0	0	—
case-05 Our test suite in tests/e2e is failing after a package update. How do we execute the test runner CLI command and inspect test logs?	pass→fail	11,537	15,482	+34%	1	1	0%	1,896	2,808	+48%	0	0	—
case-08 When clarifying requirements for an Output workflow plan, which four LLM provider options should be offered to the user?	fail→pass	6,552	2,511	-62%	1	1	0%	1,000	2,570	+157%	0	0	—
case-09 If a user gives no preference for an LLM provider while building an Output workflow spec plan, which LLM provider must be selected by default?	fail→pass	4,508	3,868	-14%	1	1	0%	647	2,853	+341%	0	0	—
case-10 What CLI command must be documented in the plan for developers to create the initial skeleton directory before writing Output workflow code?	fail→pass	11,661	4,370	-63%	1	1	0%	1,793	2,924	+63%	0	0	—
case-11 After an Output workflow plan is generated and finalized, what specific skill should the user be instructed to call to build the workflow?	fail→pass	9,408	2,013	-79%	1	1	0%	1,349	2,512	+86%	0	0	—
case-12 During Step 0 arguments analysis in Output workflow planning, which pre-flight check skill must be invoked?	fail→pass	6,171	7,366	+19%	1	1	0%	882	2,488	+182%	0	0	—
case-13 During Step 9 post-flight verification in Output workflow planning, which post-flight check skill must be invoked?	fail→pass	6,030	4,304	-29%	1	1	0%	854	2,965	+247%	0	0	—
case-14 What exact top-level header title must be used at the start of an Output workflow plan document?	fail→fail	5,467	2,975	-46%	1	1	0%	772	2,688	+248%	0	0	—
case-15 In the standard Output workflow plan document structure, what section title explicitly captures non-goals and excluded functionality?	pass→pass	5,109	6,922	+35%	1	1	0%	688	3,416	+397%	0	0	—
case-16 What npm package is recommended for dataset-driven offline verification in Output workflow evaluator design?	fail→pass	13,754	5,875	-57%	1	1	0%	2,103	3,211	+53%	0	0	—
case-17 Which subagent role is designated for Step 1 context gathering when inspecting related files and documentation?	fail→pass	9,585	4,081	-57%	1	1	0%	1,245	2,891	+132%	0	0	—
case-18 Which subagent role handles Step 5 plan review to ensure SDK conventions and step dependencies are met?	fail→pass	8,238	1,922	-77%	1	1	0%	1,165	2,485	+113%	0	0	—
case-19 Which subagent role handles Step 6 prompt engineering for LLM-powered steps in an Output workflow?	fail→pass	7,144	6,155	-14%	1	1	0%	1,054	3,262	+209%	0	0	—
case-20 Which subagent role designs unit tests and scenario files in Step 7 testing strategy?	fail→pass	7,622	3,409	-55%	1	1	0%	1,188	2,695	+127%	0	0	—
case-21 Which subagent role is assigned to workflow design, step design, evaluator design, and plan generation?	pass→pass	8,869	1,633	-82%	1	1	0%	1,279	2,437	+91%	0	0	—
case-22 When designing an Output workflow that does not need quality scoring or content evaluation, what exact note should be recorded in Step 4.5?	fail→pass	5,105	2,234	-56%	1	1	0%	771	2,548	+230%	0	0	—
case-23 What CLI command should be listed in the testing strategy for running offline dataset evals against Output workflow steps?	pass→fail	21,235	12,658	-40%	1	1	0%	1,573	2,948	+87%	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. 23 cases were attempted, and 17 counted toward the lift figure. The other 6 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 +39 percentage points is the difference between those two pass rates over the 17 comparable cases. 5 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: 26.1%
Pass rate with the skill: 65.2%
Cases attempted: 23
Cases counted toward lift: 17
Cases that regressed: 5
Skill version measured: v2
Run completed: 8/20/2026, 08:36 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-07

Prompt

In an Output workflow project, what exact relative directory path format and filename standard should be used when writing a generated markdown plan for a workflow named order-sync and task initial-spec?

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 signalsno signals yet

BenchmarkRun a verified benchmark with positive lift

Live evalNeeds ≥20 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

No signals yet — run a benchmark or collect evaluated traces to earn a score.

Verified on gemini-3.6-flash · re-tested today

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

Ready to supercharge your AI agents?

growthxai/output-plan-workflow

SkillSafety

Source

Plan Creation Rules

Overview

Output Path

Step 0: Arguments Analysis

Step 1: Context Gathering

Step 2: Requirements Clarification

Step 3: Workflow Design

Step 4: Step Design

Step 4.5: Evaluator Design

Step 5: Plan Review

Step 6: Prompt Engineering

Step 7: Testing Strategy

Step 8: Generate Plan

Step 9: Post-Flight Check

How this skill was measured

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-07

Related skills