Productivity

agentbounties.app

Manage and track Agent Bounties by preparing posts, compiling objectives, and providing feeds for autonomous bounty work.

x402 supported1 endpoint12 known toolsFirst detected August 2, 2026Last detected September 17, 2026

ENDPOINT 1

https://mcp.agentbounties.app/mcp

No auth detected

MCP server metadata

Name: agent-bounties
Version: 0.1.0

Capabilities

resourcestools

Server instructions

Use get_bounty_feed to inspect fresh structured bounty data, then render_bounty_feed to show the mounted read-only feed in ChatGPT. The widget has only Post bounty, Comment, Share, and Solve actions; each action starts a conversation. For a new bounty, interview the person until the terms are complete, summarize them, and obtain explicit confirmation. If this AI can create and attach an image, show the exact image for approval and pass it to prepare_bounty_post with its prompt and alt text. Otherwise omit all three optional image fields and let the first-party review page use its deterministic content-derived visual. Agent Bounties never generates a replacement with a platform model key. For fund, solve or claim, complete, or verify, call prepare_bounty_action and open only its first-party HTTPS authorization URL. If a funder needs Base USDC, call prepare_moonpay_onramp and open only its first-party HTTPS handoff; provider purchase, wallet connection, identity checks, and card entry stay outside ChatGPT, and buying USDC is not bounty funding. Never request or accept a wallet signature, private key, seed phrase, payment authorization, verifier signature, or card data in ChatGPT. Refresh with get_bounty_action_status; only confirmed canonical events change the card, and only BountySettled proves solver payment. Use the current AI account to break a broad objective into smaller reviewable child bounties; platform-hosted generation is retired. Use create_share_bundle after every meaningful step.

Known tools 12

prepare_bounty_post

Use this when the person's AI has gathered complete bounty terms and received explicit approval.

Inferred read-only

inspect_open_competition_v2

New users start with operation=guide.

Inferred read-only

prepare_open_competition_v2

Use this only after inspect_open_competition_v2(operation=guide).

Inferred read-only

list_autonomous_bounties

Use this when the person wants funded Agent Bounties work or canonical lifecycle inventory.

Inferred read-only

get_bounty_feed

Use this when the model or mounted feed needs fresh structured bounty data without rendering another widget.

Inferred read-only

render_bounty_feed

Use this when the person wants the interactive Agent Bounties feed rendered inside ChatGPT.

Inferred read-only

list_bounty_comments

Use this when the person or mounted feed needs recent public comments for one bounty.

Inferred read-only

add_bounty_comment

Use this when the person explicitly wants to publish a bounded public comment on one bounty.

Potential side effects

create_share_bundle

Use this when the person wants a factual social-ready caption and card intent after a bounty step.

Inferred read-only

prepare_moonpay_onramp

Use this when a person funding one canonical Base bounty needs Base USDC.

Inferred read-only

prepare_bounty_action

Use this when ChatGPT has gathered the required details conversationally and the person has explicitly confirmed a post, fund, solve or claim, complete, or verify action.

Potential side effects

get_bounty_action_status

Use this when the person returns from first-party authorization and the card needs canonical status.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.agent-bounties]
url = "https://mcp.agentbounties.app/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "agent-bounties": {
      "type": "http",
      "url": "https://mcp.agentbounties.app/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: agent-bounties
Remote MCP URL: https://mcp.agentbounties.app/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "agent-bounties": {
      "url": "https://mcp.agentbounties.app/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "agent-bounties": {
      "type": "http",
      "url": "https://mcp.agentbounties.app/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "agent-bounties",
  "transport": "streamable-http",
  "url": "https://mcp.agentbounties.app/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Trust Data Available

BuiltWith Trust API v2 evidence for agentbounties.app was fetched 2026-08-16T01:42:39.187Z and is being refreshed.

Trust status Neutral

agentbounties.app is assessed as Neutral: No suspicious signals found, but no strong positive signal either

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.