Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    CognitiveMirror-Jev-MCP

    1

    A four-tier agent orchestration package for intent mirroring, Jev profiling, decision support, and governance.

    Free

    Security scanned
    CognitiveMirror-Jev-MCP

    CognitiveMirror-Jev-MCP

    Example session with this skill installed

    I need to draft a message to my hiring manager about the interview yesterday. I am worried I sounded too technical and want to focus on my leadership experience instead.

    • Read your context and instructions
    • Compiled the cognitivemirror-jev-mcp

    The request has been processed through the mirroring and profiling tools. A decision framework and draft strategy are provided below.

    Intent Mirroring & Decision Framework

    AttributeValue
    Primary IntentDraft interview follow-up email
    ConstraintShift focus from technical skills to leadership
    Jev ProfileSoftware Engineer, Late-stage candidate, High-anxiety
    Response Modeact_with_confirmation
    Risk LevelLow (No PII or regulated advice detected)
    Business ImpactTarget: Advance to offer stage (Priority: 85/100)
    ROI HypothesisHigh-impact follow-up increases offer probability by 15%

    Recommended Action Strategy

    1. Acknowledge the technical depth of the interview as a foundation for team leadership.
    2. Pivot to leadership by citing specific examples of project management and mentoring.

    Address the "unfocused" concern by tying both technical and leadership skills into a unified "Lead Engineer" value proposition.

    Guardrail Status

    • PII Check: Passed (Redactions applied to candidate name)
    • Policy Profile: default
    • Approval Required: Yes (High-impact communication)

    Next steps

    • Review the drafted leadership pivots in the generated email body.
    • Call record_feedback once the hiring manager responds to update the Jev journey stage.
    • Invoke memory_store with consent: true to save this stakeholder interaction for future follow-ups.

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

    About this skill

    Agents often act on raw text without understanding business intent or checking for sensitive data. This package provides a four-tier architecture for intent mirroring, persona profiling, decision framing, and automated governance. It converts vague user requests into executable, risk-assessed strategies.

    What it does

    • Intent Mirroring converts raw messages into structured goals, constraints, and response strategies like clarify_first or escalate.
    • Persona Profiling builds Jev profiles using journey stages and value drivers while automatically stripping protected attributes.
    • Decision Support frames options with weighted scores, risk registers, and ROI hypotheses for measurable business impact.
    • Automated Governance checks for PII, secrets, and regulated advice flags before any high-impact action is taken.

    How it works

    1. Call mirror_user_intent first to identify the core request, urgency, and necessary response mode.
    2. Invoke build_jev_profile to retrieve stakeholder context and journey-stage alignment.
    3. Pass intent and profile data to recommend_next_action to generate instructions and impact metrics.
    4. Run check_guardrails on the final output to ensure compliance and detect data leaks before execution.

    Frameworks & tools

    This MCP package exposes a REST API compatible with any LLM orchestration layer. It includes specialized support for the Jev profiling engine and optional hybrid processing via TYPESAFE_API_KEY.

    Why this beats prompting it yourself

    Prompting an agent to be "safe" or "smart" often fails under edge cases or complex business logic. This skill enforces a deterministic workflow that separates understanding from execution, ensuring every action is audited against policy profiles like strict or enterprise.

    Use cases

    • Sales Enablement by profiling stakeholders and mapping value-drivers to specific deal stages.
    • Customer Support by detecting crisis language and escalating high-risk interactions automatically.
    • Enterprise Automation by redacting PII and enforcing governance during high-impact data workflows.
    • Executive Coaching by framing decisions with weighted tradeoffs and success metrics.

    Known limitations

    Forgetting subject memory requires admin role. Protected attributes are ignored during profiling. Hybrid mode requires a valid Typesafe API key.

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

    Requires admin role for full memory wipes. Protected attributes are ignored by design. Hybrid profiling mode requires an external API key.

    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

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 18 days ago

    • Free with an Agensi account

    Accounts
    GitHub
    Software
    PythonCognitiveMirror Jev MCP001 MCP002 MCP003 MCP004 MCP005 MCP006 MCP007 MCP008 MCP

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files
    Environment Variables

    Allowed Hosts

    api.github.com
    Listed18 days ago

    What's inside

    Frequently Asked Questions

    This skill solves the problem of 'hallucinated intent' and governance gaps by enforcing a four-step workflow—mirroring, profiling, framing, and auditing—ensuring agents act on validated business logic rather than just raw text.

    The package includes the core orchestration engine, pre-configured policy profiles (like 'strict' and 'enterprise'), and the Jev profiling schema, delivered as a REST API-compatible MCP package.

    It is designed to be LLM-agnostic and works with any orchestration layer (like LangChain, AutoGPT, or custom frameworks) that can call REST APIs; hybrid processing requires a 'TYPESAFE_API_KEY'.

    Setup involves deploying the MCP server and integrating the four specific tool calls—intent mirroring, profiling, decision framing, and guardrails—into your agent's existing execution loop.

    Yes, while the system automatically strips protected attributes for compliance, users with 'admin' roles can manage subject memory and adjust the sensitivity of the PII and secret detection guardrails.

    Buyers receive lifetime access to version 1.x updates, including security patches for the guardrail engine and optimizations for the Jev profiling logic.

    New in LLM Ops

    Free