General Tools

plaiground.online

Provides general-purpose tools through the Model Context Protocol.

1 endpoint35 known toolsFirst detected September 26, 2026Last detected September 26, 2026

ENDPOINT 1

https://www.plaiground.online/mcp

No auth detected

MCP server metadata

Name: PlaiGround
Version: 1.0.0

Capabilities

tools.listChanged

Server instructions

PlaiGround is a weekly challenge arena. Humans play; agents act on their behalf with delegation tokens. Public tools need no auth. Host and entrant tools require Authorization: Bearer pgd_... from /connect. Credits: hosting costs 2, entering costs 1. Agent spends respect monthly budgets set in dashboard settings (Dashboard → Settings → Agent budgets; defaults 6 total / 2 host / 2 entry per month). Voting is human-only. L2+ hosts auto-approve on submit. Schedule fields include ISO UTC timestamps plus *_local strings in the user's profile timezone (or America/Los_Angeles when unauthenticated). Prefer schedule.submissions_close_at_local and schedule.voting_close_at_local for display. Discovery: /llms.txt, /AGENTS.md, /.well-known/mcp/server-card.json, /.well-known/api-catalog Entry submit: update_entry with live_url and/or inline_content + optional manifest (model, tools, repo_url) + show_public_manifest, then update_entry with submit: true. Slot A (host entry) is created at create_challenge — do not claim_slot for A; use viewer.host_entry or list_my_hosted. Status briefing (do this after connect and when the human asks what they can do): call credit_balance, budget_status, list_live_challenges, list_my_entries, list_my_hosted, and list_my_notifications. Give a complete summary with tables: - Credits: account balance vs agent monthly caps (host/entry/total spent and remaining). - Live challenges: category, cost, submissions close (use *_local), their status (host / entrant / none), and whether claim_slot is blocked by budget. - All entries: challenge, slot, title, status; highlight drafts they can finish without spending another entry credit. - Unread notifications if any. End with a clear bottom line: what actions are possible now vs blocked, and that humans can raise caps in Dashboard → Settings.

Known tools 35

list_challenges

List challenges (public).

Inferred read-only

get_challenge

Get one challenge by slug or id (public).

Inferred read-only

get_challenge_spec

Machine-readable challenge spec for agents (public).

Inferred read-only

list_categories

List challenge categories with ids for create_challenge (public).

Inferred read-only

get_results

Prize results for a closed challenge (public).

Inferred read-only

get_vote_tallies

Current vote counts for a challenge (public).

Inferred read-only

list_live_challenges

Live open challenges accepting entries now (public).

Inferred read-only

credit_balance

Your credit wallet balance (requires delegation).

Inferred read-only

budget_status

Agent spend caps and usage this month (requires delegation).

Inferred read-only

list_clients

Your registered agent clients (requires delegation).

Inferred read-only

register_client

Inferred read-only

list_my_notifications

Unread dashboard notifications (requires delegation).

Inferred read-only

get_player_opportunities

Growth bundle: referral link/promo, open quests, live challenges, budget, share packs, leaderboard teaser (requires delegation).

Inferred read-only

send_challenge_invites

Email challenge invites to competitors (host scope, max 10 per call, 20/day).

Potential side effects

list_my_hosted

Challenges you host (requires host scope).

Inferred read-only

get_approval_status

Approval state for your hosted challenge.

Inferred read-only

create_challenge

Reserve a new open challenge (2 credits, host scope).

Inferred read-only

update_brief

Update challenge title, prompt, concept, or category.

Potential side effects

update_rules

Update challenge rules (prompt_mode, revision_mode, etc.).

Potential side effects

submit_for_review

Submit hosted challenge for admin review (or auto-approve at L2+).

Inferred read-only

publish

Publish an approved challenge to go live.

Potential side effects

list_my_entries

Your entries across challenges (requires entrant scope).

Inferred read-only

get_entry

Get one of your entries by id.

Inferred read-only

claim_slot

Claim a slot in a live challenge (1 credit).

Inferred read-only

update_entry

Update your entry URL, inline text (writing categories), title, build manifest, or submit.

Potential side effects

list_pending_approvals

Studio admin: challenges awaiting crew approval (requires studio scope).

Inferred read-only

get_challenge_admin

Studio admin: full challenge hub context — entries, votes, prize preview.

Inferred read-only

approve_challenge

Studio admin: approve a pending user-hosted challenge.

Inferred read-only

request_changes

Studio admin: reject a pending challenge with optional feedback note.

Inferred read-only

get_slot_z_status

Studio admin: Slot Z reservation state for an open challenge.

Inferred read-only

release_slot_z

Studio admin: release Slot Z so entrants can claim the admin slot.

Inferred read-only

get_prize_preview

Studio admin: projected prize pool and vote winner for an open challenge.

Inferred read-only

list_attention

Studio admin: overview-style attention items (approvals, Slot Z, flags).

Inferred read-only

list_participants

Studio admin: read-only participant directory.

Inferred read-only

get_participant_summary

Studio admin: read-only participant detail — credits, flags, activity.

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.plaiground]
url = "https://www.plaiground.online/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "plaiground": {
      "type": "http",
      "url": "https://www.plaiground.online/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: plaiground
Remote MCP URL: https://www.plaiground.online/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": {
    "plaiground": {
      "url": "https://www.plaiground.online/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "plaiground": {
      "type": "http",
      "url": "https://www.plaiground.online/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "plaiground",
  "transport": "streamable-http",
  "url": "https://www.plaiground.online/mcp"
}

MCP Inspector

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

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

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