Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    personal curate document skill

    1

    Standardized one-line task tracking and document templates for token-lean, agent-ready project documentation.

    $6

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    personal curate document skill

    personal curate document skill

    Example session with this skill installed

    Update the Build Plan for the Auth module. We finished the JWT implementation today, but the Refresh Token logic is still pending because the database schema isn't ready.

    • Read your context and instructions
    • Compiled the personal curate document

    I have updated the Auth module build plan to reflect the completion of the JWT implementation and the current blocker for Refresh Token logic.

    Build Plan

    Auth Module

    1. JWT Implementation — started 2024-05-20 — Secure token generation and header validation — closed 2024-05-22
    2. Refresh Token Logic — started 2024-05-22 — Persistent session management and rotation —

    pending (remaining: implement rotation logic and verify against db)

    • (i) Blocked by pending migrations in the users table.

    State today — 2024-05-22

    1. Core authentication via JWT is now functional and closed.
    2. Work has shifted to session persistence, currently waiting on database schema updates.

    Next steps

    1. Finalize the database schema for the refresh_tokens table.
    2. Complete the rotation logic once migrations are applied.

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

    About this skill

    Technical documentation often fails because it mixes status, intent, and execution into long, unparseable paragraphs. This skill enforces a strict, one-line task grammar that ensures every project document remains a source of truth for both humans and AI agents. It eliminates the "where was I?" problem by mandating a dated state snapshot and specific metadata for pending, parked, and closed tasks.

    What it does

    • Enforces task grammar by requiring a specific format for every task: name, start date, summary, and status.
    • Manages task lifecycle using three distinct states: closed, pending (with remaining work details), and parked (with owner-hold notes).
    • Structures project docs via a fixed six-section menu (Direction, State Today, Build Plan, etc.) to ensure uniform scannability.
    • Captures technical debt through (i) notes that document "hard parts" directly under the relevant task line.
    • Generates fitted templates for Roadmaps, Specs, MVPs, and Feature docs based on the specific scope of work.

    How it works

    1. Initialize a new doc by requesting a fitted template (ROADMAP, SPEC, MVP, or FEATURE).
    2. Read existing docs in a specific order to ingest direction, current state, and pending hard parts without re-deriving context.
    3. Record work by updating task lines in the same turn the work is completed, following the one-line grammar.
    4. Transition status by moving tasks to closed (with dates), pending (with remaining items), or parked (with owner timestamps).

    Frameworks & tools

    This skill is framework-agnostic and works with any Markdown-based documentation system, including GitHub wikis, Notion, and local project directories.

    Why this beats prompting it yourself

    Generic prompts allow agents to write verbose, drifting updates that eventually bloat the context window. This skill enforces token-lean documentation that prevents "hallucinated progress" by requiring specific metadata for every line.

    Use cases

    • Resuming a complex coding project after a long break without losing technical context.
    • Maintaining a shared roadmap that both developers and agents can update programmatically.
    • Drafting technical specifications where open questions and blockers are clearly isolated.
    • Creating a lean MVP build plan that tracks exactly what is remaining for a shippable slice.

    Known limitations

    Does not apply to README files, which remain factually current only. Task summaries are strictly limited to 20 words or less.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Technical Documentation.

    Summaries must be 20 words or less. Not for README files. Requires YYYY-MM-DD date format. Plain closed lines cannot carry remaining work notes.

    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 6 days ago

    Permissions

    Read Files
    Listed6 days ago

    What's inside

    Frequently Asked Questions

    This skill addresses "context bloat" and "documentation drift" by enforcing a strict, one-line grammar for tasks, ensuring both humans and AI agents can instantly parse project status without reading through long, verbose paragraphs.

    The skill is designed for any Large Language Model (LLM) or AI agent that supports Markdown, and it integrates seamlessly with documentation hosted on GitHub, Notion, or local development directories.

    Each purchase includes the core logic for the one-line task grammar, the six-section project menu structure, and four specialized templates for Roadmaps, Specs, MVPs, and Feature documents.

    Yes, it is designed for multi-agent workflows; the standardized grammar ensures that Agent B can pick up exactly where Agent A left off by reading the 'State Today' and 'Pending' task metadata.

    The skill is framework-agnostic and relies on standard Markdown, meaning it requires no heavy installation and can be integrated into your agent's workflow immediately upon deployment.

    To maintain token efficiency and scannability, task summaries are strictly limited to 20 words or less, and it is not intended for README files which serve a different purpose.

    New in Technical Documentation