Honest Plan Diagram

    1

    Turn a prompt into a reviewable agent plan diagram before any subagent starts. The graph is the contract the agent will actually run, not a caption of work happening somewhere else.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Honest Plan Diagram

    Example session with this skill installed

    I need to research the top 5 competitors in the AI agent space, summarize their pricing, and draft a comparison email for my team. Show me the plan first.

    • Read your context and instructions
    • Compiled the honest plan diagram

    I have interpreted your request as a multi-stage research and drafting operation requiring competitive intelligence gathering, data synthesis, and a final communication draft.

    Proposed Research and Drafting Plan

    graph TD
        subgraph web-research [Web Research]
        node-1[Search Competitors]:::source
        node-2[Extract Pricing]:::source
        end
    
        node-3[Normalize Data]:::transform
        node-4[Compare Features]:::decision
        node-5[Store Findings]:::store
        
        subgraph communication [Communication]
        node-6[Draft Comparison Email]:::transform
        node-7[Human Review]:::human_gate
        node-8[Send to Team]:::transform
        end
    
        node-1 --> node-3
        node-2 --> node-3
        node-3 --> node-4
        node-4 --> node-5
        node-5 --> node-6
        node-6 --> node-7
        node-7 --> node-8
    
        classDef source fill:#f9f,stroke:#333,stroke-width:2px
        classDef transform fill:#bbf,stroke:#333,stroke-width:2px
        classDef decision fill:#ff9,stroke:#333,stroke-width:2px
        classDef human_gate fill:#f96,stroke:#333,stroke-width:2px
    

    What waits on you:
    The Human Review gate will pause execution after the email is drafted but before it is sent. You will be able to edit the draft or change the recipient list.

    Note: No research has started and no tools have been called. I am standing by for your approval to initialize this plan.

    Next steps

    1. Type Approve to begin the search and data normalization.
    2. Type Move Review to shift the human gate earlier (e.g., after data normalization).
    3. Type Edit Plan to add additional competitor targets.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Most agents start working the moment you hit send. You get a spinner, then a surprise. This skill makes the agent draw the plan first. It maps the job onto a small, fixed vocabulary: source, transform, store, decision, human gate, and loop. Parallel work is grouped into branches you can later pause or kill. Anything that would email, post, or submit has to pass a human gate. Trivial requests stay small. You get a one-paragraph read of the goal, a Mermaid map of the proposed graph, a chance to move the review gate, and a hard stop until you approve. A bundled checker validates the plan JSON and prints the diagram. Nothing talks to a paid API. It runs inside whatever agent you already have. Where it stops: it does not execute the plan. After you approve, execution is on you (or on the companion skills).

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.

    No support for cycles; loops must be metadata-defined. Requires Python 3 for the validator script. Designed for multi-step tasks; skips complexity for simple single-action prompts.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 19 days ago

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed19 days ago

    What's inside

    Frequently Asked Questions

    It prevents "black box" behavior by forcing the agent to output a validated JSON plan and Mermaid diagram for your approval before it executes a single tool or external action.

    The skill is designed for any multi-agent framework that supports JSON state management and includes a Python-based validator script to ensure the agent's logic is structurally sound.

    You receive the core planning logic, the Mermaid.js visualization template, and the `validate_plan.py` script used to audit the agent's proposed DAG for errors.

    No, the skill is designed to prevent cycles; loops must be defined within the node metadata to keep the execution graph clean and readable.

    While it uses Mermaid.js for rendering, the skill includes a rigid ontology of six specific node types (Sources, Transforms, etc.) to ensure the plan is a functional contract rather than just a pretty picture.

    Updates are handled through the marketplace dashboard, where you can download new versions of the validation script and schema as they are optimized for newer LLM models.

    New in Agents & Orchestration