arc-market-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| ARC_API_BASE | No | Base URL for the `/arc/*` routes. | https://api.babacapital.app |
| ARC_API_TIMEOUT_MS | No | Per-request timeout in milliseconds (the sellsim route runs a live simulation). | 20000 |
| ARC_AGENT_CARD_BASE | No | Base URL for `/.well-known/agents/<name>.json`. | https://babacapital.app |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| arc_healthA | Liveness of the Arc Data API and the list of routes it serves. Returns the §3.5 envelope. Call this first to confirm the API base and see which live routes are wired. |
| arc_launchpadsA | Launchpad concentration on Arc over the recent window: HHI, pools created, v3 vs v4 launch counts, distinct deployer callers, and per-launchpad share/graduations/dead-on-arrival. NO_DATA until an hourly reading has been written; partial windows are NO_DATA, never a false zero. |
| arc_lp_aprA | Realised range-APR for Arc LP pools from the latest scan. Optional filters: pair (e.g. 'usdc/weth'), venue (e.g. 'aero-lite'), range_pct. APR assumes price stayed in range over the window; the warning states this. NO_DATA if no scan has been written. |
| arc_rpc_consensusA | How many of Arc's four public RPCs agreed at the last reading (of 4), and the head they agreed on. Below 2 the status is REFUSED — treat any Arc figure as not cross-checked until they agree again. A cheap trust gate before acting on other Arc data. |
| arc_token_screenA | Screen a launched Arc token by contract address: its USDC-paired pool, fee, recent swap count, and the decoy-pool verdict (a fee above the cap, or a dynamic pool whose swaps charged above the cap, is refused). A cheap gate — it does NOT run a live buy/sell. NO_DATA if the token has not traded in the API's discovery window. |
| arc_token_sellsimA | Honeypot test for an Arc token: a LIVE sell-simulation (buy then sell in one simulated transaction against current chain state) reporting the round-trip fraction recovered. Below ~0.9 the token is likely a honeypot or a punitive-tax trap. This is a heavy, paid-tier route. NO_DATA if the pool cannot be simulated (it is never scored zero on a failed simulation). |
| arc_pool_decoyA | Decoy-pool verdict for a specific v4 poolId: whether this pool is the real venue for its token or a high-fee decoy that a naive buyer would route into. Explains the refusal reason. NO_DATA if the poolId is unknown to the API. |
| arc_agent_cardB | Fetch a BABA Capital ERC-8004 agent-card (the /.well-known/agents/.json identity document). Known names: 'baba-arc-data', 'baba-arc-yield'. Shows the agent's registry, agentId (null until registered), and x402 payment network. |
| arc_agents_listA | List the ERC-8004 agents registered on Arc, from a validated hourly crawl. Each row carries agentId, owner, agentURI, host, name, skills, status (OK | UNRESOLVED | INVALID), whether it declares x402, and how concentrated its host and owner are. Optional filters: status, host, owner. Paged. An agent whose card will not resolve is listed as UNRESOLVED, never silently dropped. |
| arc_agent_getA | Full record for one Arc agent by agentId: its on-chain owner and agentURI, the RAW card as served, and a validation report showing the evidence behind every flag (which key supplied the name, which supplied the skills, the exact key and value that prove an x402 or payment declaration, redirects, robots handling, and any fetch error). Use this to check the directory's claims yourself rather than trusting them. |
| arc_agents_searchA | Find an Arc agent that does something, and optionally one you can pay. |
| arc_agents_statsA | Shape of Arc's ERC-8004 agent registry in one call: total ids, distinct owners, distinct hosts, how many resolve, how many declare x402 vs payment generally, status counts, and the most concentrated hosts and owners with their share of the registry. Facts only — the concentration numbers are stated so you can draw your own conclusion. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 12 tools
Most tools are clearly distinct: health, launchpads, LP APR, RPC consensus, token screening/simulation, pool decoy, and agent directory operations each target a different resource. The only mild overlap is arc_agents_list vs arc_agents_search vs arc_agent_get, but their descriptions make the distinction clear (list all, search, get one).
All tools use the arc_ prefix with a noun or noun_verb pattern: arc_health, arc_launchpads, arc_lp_apr, arc_rpc_consensus, arc_token_screen, arc_token_sellsim, arc_pool_decoy, arc_agent_card, arc_agents_list, arc_agent_get, arc_agents_search, arc_agents_stats. The pattern is consistent, though arc_token_sellsim is slightly less readable than arc_token_sell_sim would be.
12 tools is well within the ideal 3-15 range. The set covers two coherent domains (Arc DEX data and ERC-8004 agent registry) without feeling bloated; each tool has a specific purpose and the count feels appropriate for the server's scope.
The DEX side covers health, launchpads, LP APR, RPC consensus, token screening, sell-simulation, and decoy-pool detection — a solid analytical surface. The agent side covers list, get, search, and stats, which is complete for a read-only directory. Minor gaps: no tool for historical LP APR or token price/quote, and no agent registration/update (though that may be out of scope for a read-only MCP).