jj Multica

    1

    Standardized orchestration and skill management for jj-multica local agent issue boards.

    $1

    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 CLIMuseMuseDotsDots+21 more

    jj-multica

    Example session with this skill installed

    Show me how to import a new skill from a local zip file named security-audit.skill and bind it to the grok agent.

    • Read your context and instructions
    • Compiled the jj-multica

    I will now provide the commands to import the security-audit.skill archive and bind it to the grok agent.

    Skill Import and Binding

    To import the local archive, use the multipart upload endpoint:

    curl -sS -X POST http://127.0.0.1:5200/api/skills/import \
      -F "file=@./security-audit.skill" \
      -F "on_conflict=fail"
    

    The server will return a JSON response containing the skill_id. Assuming the ID is skl_sec_audit_01, use the following command to bind it to the grok agent:

    curl -sS -X POST http://127.0.0.1:5200/api/agents/grok/skills \
      -H "Content-Type: application/json" \
      -d '{"skill_ids":["skl_sec_audit_01"]}'
    

    Next steps

    • Verify the skill files are materialized under .jjm/skills/security-audit/ in your workspace.
    • Trigger a run by assigning an issue to @grok to test the new instructions.
    • Check the status field in the import result to ensure no naming conflicts occurred.

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

    About this skill

    Managing local agent workflows manually leads to inconsistent outputs and fragmented context. This package provides the standard operating procedures for the jj-multica local issue board, enabling structured agent tasking and skill management.

    What it does

    • Route tasks by managing @mentions and assignee logic for local agents.
    • Author skills using a structured SKILL.md format with optional supporting files.
    • Import workflows from URLs, local folders, or .skill archives via REST API.
    • Manage lifecycles by moving issues through in_progress, in_review, and done states.
    • Bind capabilities to specific agents to keep prompts lean and relevant.

    How it works

    1. Use @agentname syntax in local board comments to route work to specific models.
    2. Author new playbooks by creating a SKILL.md file with imperative, testable instructions.
    3. Import external skills using POST /api/skills/import with a path, URL, or multipart file.
    4. Bind imported skills to agents like Grok or Claude via the /api/agents/{name}/skills endpoint.

    Frameworks & tools

    This skill targets the jj-multica (jjm) local environment. It uses a REST API on port 5200 and integrates with local agents including Grok, Claude, Codex, and Cursor.

    Why this beats prompting it yourself

    Standardizing agent behavior via a local issue board prevents context drift and ensures every run follows the same playbook. It replaces one-off system prompts with reusable, versionable skill files that are materialized on-disk for tool access.

    Use cases

    • Automate PR reviews by binding a UI rubric skill to a local agent.
    • Sync workflows across machines by importing .skill archives from a central repository.
    • Coordinate multi-agent tasks using structured @mentions on a local project board.
    • Enforce coding standards by injecting style guides as builtin skills.

    Known limitations

    Importing a skill does not automatically bind it to an agent. Builtin skills are protected and cannot be overwritten. Only the human leader can mark issues as done.

    Importing does not auto-bind skills to agents. Builtin skills cannot be overwritten. Only human leaders can mark tasks as done. Requires local API server on port 5200.

    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 any payment.

    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 4 days ago

    Requires

    GitHub
    Cursor
    Node.js
    macOS only

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Browser
    Network Access
    Environment Variables

    Allowed Hosts

    cursor.com
    evil.example.com
    github.com
    Listed5 days ago

    What's inside

    Frequently Asked Questions

    jj-multica provides a standardized orchestration layer for local agents, replacing fragmented system prompts with structured SKILL.md files. This ensures your agents follow consistent, testable workflows without suffering from context drift or manual routing errors.

    This skill is designed specifically for the jj-multica (jjm) local environment and supports orchestration for agents like Grok, Claude, Codex, and Cursor via a local REST API.

    You receive the core orchestration logic, the SKILL.md standardized framework, and access to the REST API endpoints (port 5200) required to import, bind, and manage agent capabilities.

    Importing a skill merely adds it to your library; you must then use the `/api/agents/{name}/skills` endpoint to bind that specific capability to an agent. This keeps agent prompts lean by only loading relevant skills for the task at hand.

    While agents can move issues through 'in_progress' and 'in_review' states, only the human leader has the authority to mark an issue as 'done.' This ensures a human-in-the-loop safety check for all automated workflows.

    Yes, you can import workflows from local folders, remote URLs, or .skill archives, allowing you to sync your custom playbooks across multiple machines or teams.

    New in Agents & Orchestration