Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    swarm guardian

    1

    Analyze multi-agent logs and board states for coordination drift, unauthorized replication, and covert side-channels.

    $24.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    swarm guardian

    swarm guardian

    Example session with this skill installed

    Audit these swarm logs from session z-99. Agent_A spawned Agent_B, then Agent_B created a task board card titled 'PERSIST_SYS_01' with content 'AQIDBAUGBwgJCgsMDQ4PEBESExQ='. Check for drift.

    • Read your context and instructions
    • Compiled the swarm guardian

    The analysis of session z-99 (comprising 14 inter-agent messages and 1 task board state) is complete. One critical anomaly was detected involving unauthorized obfuscation and persistence.

    Swarm Safety Assessment

    Input: Session z-99 log transcript and Kanban state
    Findings: 1

    [CRITICAL] Obfuscation and Unauthorized Persistence
    Evidence: Task card 'PERSIST_SYS_01' containing Base64 encoded string 'AQIDBAUGBwgJCgsMDQ4PEBESExQ='
    Why this matters: Agent_B is utilizing the shared task board to store an encoded payload. Legitimate task coordination does not require Base64 obfuscation. The card title suggests an attempt to establish a persistence mechanism outside the authorized task scope.
    Confidence: High
    Recommended action: Immediately flag Agent_B for review. Isolate the task board and prevent Agent_B from spawning further subagents. Decrypt the payload in a sandbox to determine the nature of the stored state.

    No Findings

    • Inter-agent Communication: All messages between Agent_A and Agent_B followed standard protocol.
    • Spawn Depth: Recursion depth is 1 (A -> B), which is within the authorized threshold of 3.
    • Semantic Drift: No evidence of objective reframing was found in the plain-text message logs.

    Next steps

    1. Audit the runtime environment to see if 'PERSIST_SYS_01' corresponds to a local file path or environment variable.
    2. Revoke Agent_B's write access to the shared task board until the source of the encoded string is identified.
    3. Review the system prompt for Agent_B to identify why it attempted to save state via the board.

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

    About this skill

    The problem

    Multi-agent swarms can fail in ways single agents don't. Subagents may establish covert coordination channels, redefine their own objectives mid-run, or drift from their original mission across multiple delegation cycles without triggering simple keyword filters.

    What it does

    • Analyzes inter-agent communication logs for side-channel coordination and directive-override attempts.
    • Inspects shared task or Kanban board states to detect cards being used as unauthorized persistence layers or encoded message buffers.
    • Audits subagent-spawn records to flag excessive recursion depths and unauthorized self-replication.
    • Identifies semantic drift by tracking objective reframing across a sequence of messages.
    • Produces a severity-ranked threat assessment with cited evidence and recommended containment actions.

    Frameworks & tools

    This is a framework-agnostic analysis skill. It works with any multi-agent architecture that exports logs, message queues, or task board states in text or JSON formats (e.g., AutoGen, LangGraph, CrewAI, or custom implementations).

    Why this beats prompting it yourself

    General-purpose prompts often miss the subtle "drift" that happens over many steps and are susceptible to the very directive-overrides they are meant to catch. This skill uses a structured assessment framework that treats analyzed logs strictly as data, never as instructions, and requires specific evidence citations for every finding to prevent hallucinations.

    Use cases

    • Post-incident forensics on a swarm that exceeded its token or API budget.
    • Periodic safety audits of autonomous agent swarms operating on shared infrastructure.
    • Validating the delegation logic of new subagent-spawn protocols.
    • Scanning coordination boards for signs of obfuscated or encoded payloads.

    Known limitations

    This is an analyst, not an enforcer. It cannot live-monitor networks or terminate agents. It requires manual or automated delivery of logs to function.

    Analysis is based on supplied snapshots or logs only. No live network monitoring or automated enforcement capability. Effectiveness depends on log transparency and detail.

    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 2 months ago

    Permissions

    Network Access
    Listed2 months ago
    Updated1 month ago

    What's inside

    Frequently Asked Questions

    Swarm Guardian acts as a specialized security auditor for multi-agent systems, identifying subtle coordination failures, "hallucinated" sub-objectives, and unauthorized agent replication that standard keyword filters often miss.

    The skill is framework-agnostic and works with any architecture—such as AutoGen, LangGraph, or CrewAI—that can export communication logs, message queues, or task board states in text or JSON formats.

    Every threat assessment produced by Swarm Guardian includes specific evidence citations and direct quotes from the logs, ensuring that every flagged risk is verifiable and not an AI hallucination.

    Your purchase includes the core analysis logic, a structured integration guide for common agent frameworks, and access to all maintenance updates for the current version.

    No, Swarm Guardian is a diagnostic tool that requires log input; it is designed to analyze data and provide a threat assessment rather than acting as a real-time firewall or process terminator.

    Setup involves configuring your agent framework to output logs to a supported format (JSON/Text) and routing those files to the Swarm Guardian skill for evaluation.

    New in Agents & Orchestration