EthanXiang777/ash-agents

A swarm of disposable agents for unattributable web interaction

What it solves

Ash Agents solves the problem of persistent online identity and tracking. Instead of trying to hide a single identity using VPNs or private browsing, it removes the persistent identity entirely by using a swarm of disposable agents to interact with the web on your behalf.

How it works

Users provide instructions in plain language, which the system decomposes into tasks. These tasks are are assigned to agents that operate in isolated browser contexts with unique fingerprints, device characteristics, and network egress. Once a task is completed, the agent and its entire identity are destroyed, ensuring that no state is retained for websites to correlate or track.

Who it’s for

Individuals seeking high-level privacy and non-attribution when interacting with web services, specifically those who want to prevent the formation of durable user profiles based on behavioral correlation and fingerprinting.

Highlights

  • Non-attribution: Prevents interactions from being linked back to the user's own environment.
  • Non-linkability: Ensures different tasks cannot be linked to the same source because each agent uses a unique fingerprint.
  • No persistent profile: Denies the ability for services to build a long-term behavioral record by destroying identities immediately after use.
  • Private Coordination: Orchestration of the swarm happens within the user's trust boundary, so the web only sees unrelated participants.

Related

  • Project
    canyonroad/agentsh

    A secure, policy-enforced execution gateway for AI agents that intercepts and governs file, network, process, and database activity at runtime.

  • Project
    desplega-ai/agent-swarm

    agent‑swarm is an open‑source platform that turns a lead AI agent into a task‑routing hub, delegating work to containerised worker agents (Claude Code, Codex, etc.). It integrates with Slack, GitHub, email, and many other tools, persisting context via vector memory, and can automatically generate PRs, replies, or scheduled scripts. The system is model‑agnostic, runs via Docker (or Helm on Kubernetes), and is licensed under MIT.

  • Project
    Panniantong/Agent-Reach

    Agent Reach is an open‑source Python CLI that equips any LLM‑based AI agent with ready‑made, auto‑updating ways to read webpages, fetch YouTube subtitles, query Reddit, pull GitHub data, and access many other platforms. It selects, installs, and health‑checks the best open‑source tools for each service, registers a `SKILL.md` for the agent, and provides a `doctor` command to diagnose channel status. The project is free, MIT‑licensed, and focuses on safety (local credential storage, no system changes without explicit `--system`).

  • Project
    swarmclawai/swarmclaw

    A self-hosted AI agent runtime and multi-agent framework for creating autonomous agent swarms with durable memory, scheduling, and multi-provider LLM support.

  • Project
    VRSEN/OpenSwarm

    A multi-agent system that coordinates specialized AI agents to generate professional deliverables like slide decks, research reports, and videos from a single terminal prompt.