Skip to content
FlareWatch logo

FlareWatch MCP server

Which MCP servers are FlareWatch's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-15

← Back to FlareWatch

Whose server it is
  • Registry nameio.flarewatch/flarewatchverified in the MCP registry

The registry name sits under flarewatch.io. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-15
  • Stateanswerschecked 2026-10-15
  • Tools26
  • Server calls itselfflarewatch-mcp
  • Server version0.2.0
  • Protocol version2025-06-18

On 2026-10-15 the server answered initialize and tools/list without a sign-in. First answer in 314 ms.

Should an agent connect itNot proven either way yet
  • Whether its tools do the product's job has not been judged yet.

Once connected

  • For "compare validators using published scoring", the likely first call is compare_validators.
  • Parameters are untyped or undescribed in 13 of 26 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 9, the first on 2026-10-19
  • INITE MCP score96 of 100answers 100 · definitions 91 · context 95 · connects 100
  • Answered100% of days9 of 9 daily checks
  • Median first answer865 ms
  • Space in an agent's context≈ 3,630 tokensthe definitions of all its tools (26), which an agent carries on every turn
  • Tool definitionsA · 91 of 100
  • Typed results0 of 26 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a type: in 1 of 26
  • Parameters without a description: in 13 of 26

What an agent can do with it

  • not sure · track staking rewards and claimable epochs
  • not sure · generate tax records for staking reward income
  • can do · compare validators using published scoring

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers26 tools · read 2026-10-15
  • get_flare_validators

    List Flare P-Chain validators with FlareWatch composite Score, stake, fee, uptime, and capacity. Sort by score (default), stake, or apy. Pure on-chain + computed data; no editorial ranking. limit caps at 100.

  • get_validator_score

    Return the FlareWatch composite Score and full per-dimension breakdown for one Flare validator by NodeID. Each dimension carries its raw score, max, and the detail string explaining the score (e.g. 'Active outage · 3 consecutive missed…

  • get_ftso_providers

    List FTSO data providers on Flare or Songbird. Flare rows include FlareWatch composite Score; Songbird rows expose vote power, fee, and last-epoch reward rate (no score on SGB by design). Default sort: score (FLR) / vote power (SGB)…

  • get_ftso_provider_metrics

    Return detailed metrics for one FTSO data provider on Flare or Songbird: fee, reward rate, uptime, vote power, delegator count, and (Flare) the FlareWatch composite Score plus the provider's rank and percentile by score and by reward rate…

  • get_network_status

    Return current FTSO reward-epoch state for Flare (default) or Songbird: current epoch ID, when it started, how far into the epoch we are, and how many seconds remain. Source is FlareSystemsManager on-chain reads. Nulls returned when…

  • get_flarewatch_validator

    Return FlareWatch's own Flare P-Chain validator stats — same scoring algorithm as every other validator, no promotional weighting. Returns one row per FlareWatch-operated node, with self-bond, delegations, fee, uptime, expiry, and…

  • get_flarewatch_sgb_entity

    Return FlareWatch's Songbird FTSO entity's live on-chain stats: WSGB vote power, self-balance, current delegation fee, and per-WSGB reward rate from the last finished reward epoch. Source is direct on-chain reads (WNat + WNatDelegationFee…

  • get_flare_prices

    Latest FLR, SGB and XRP prices in USD and major fiats (EUR, GBP, AUD, JPY, CAD). Each asset has its OWN source (ftso oracle, or coingecko fallback) and sources can differ within one response, so check source per asset. degraded=true only…

  • get_recent_agent_actions

    Return the FlareWatch agent's recent action feed: surface, action kind (observation/propose/topup/rollback/claim/etc), outcome, and detect-to-remediate latency on the agent's own recoveries. Validator-surface entries are proposals only…

  • get_agent_capabilities

    Return the FlareWatch agent's capability surface: autonomous-vs-total playbook counts per family (validator families are propose-only — autonomous 0 by design), autonomous fraction across agent-owned surfaces, drill correctness, defects…

  • get_agent_status

    Return the FlareWatch agent's reliability scorecard from its hourly signed snapshot: autonomous fraction across agent-owned surfaces (validator families are propose-only) + per-family breakdown, recovery counts (deterministic = the node's…

  • get_network_economics

    Return Flare's monetary metrics: circulating and committed supply, total staked, staking ratio, WFLR supply, annualized inflation (rate + FLR/yr), and the FLR burn rate (daily + annualized estimate) with net supply change. Source is…

  • get_fassets_status

    Return FAssets status on Flare: for each FAsset (e.g. FXRP), its symbol, underlying asset, minted amount, and USD TVL (minted x underlying price), plus aggregate total_tvl_usd. Source is on-chain totalSupply reads valued at current…

  • get_defi_tvl

    Return total value locked (USD) across the Flare DeFi ecosystem with a per-protocol breakdown (protocol name, TVL, token count). Source is on-chain reads valued at current prices. available=false / total null when the price-dependent TVL…

  • get_chain_activity

    Return live Flare C-chain activity: transactions per second, average block time, block utilization, gas price (gwei), average fee per tx, latest block number, and all-time totals (transactions, addresses, blocks). Source is a live block…

  • get_ftso_provider_history

    Return one Flare FTSO provider's per-epoch reward-rate and Score trend over the last tracked reward epochs (up to 30, newest first). address may be the provider's entity address or its delegation address. found=false when the provider…

  • list_ftso_feeds

    List the FTSO block-latency price feeds Flare's oracle provides (e.g. FLR/USD, BTC/USD, ETH/USD) with Flare's published risk tier for each (low/medium/high/new — volatility + liquidity). Optionally filter by risk. These are the symbols…

  • get_payment_options

    Return machine-readable payment options for paid FlareWatch capacity: deposit USDT0 on Flare and the sending wallet is credited compute units (CU) that never expire. Includes the deposit address, credit rate, and balance URL…

  • simulate_delegation_rewards

    Estimate annual WFLR FTSO-delegation rewards for delegating an amount to a Flare FTSO provider (defaults to FlareWatch). Annualizes the provider's latest realized per-epoch reward rate (rate x ~104.29 epochs/yr). Returns estimated APR and…

  • compare_validators

    Side-by-side comparison of up to 10 Flare validators by NodeID. Rows returned in the order supplied (no re-ranking by FlareWatch). Each row carries score, stake, fee, uptime, capacity, APR, and stake-period expiry. Missing NodeIDs return…

  • compare_ftso_providers

    Side-by-side comparison of up to 10 FTSO providers on Flare or Songbird. Rows returned in the order supplied (no re-ranking by FlareWatch). Flare rows include the composite Score; Songbird rows expose vote power, fee, last-epoch reward…

  • verify_response

    Verify that a previously-served FlareWatch MCP response really came from this server. Pass the structured envelope WITHOUT the verification_signature field as payload, and the signature you received as signature. Returns valid=true when…

  • get_my_delegations

    Return the calling wallet's WFLR FTSO delegations and P-chain validator delegations. Requires OAuth scope wallet:read. The wallet is resolved from the OAuth token; no input arg accepts a wallet address (cross-tenancy isolation).

  • get_my_claimable

    Return what the calling wallet can claim right now: Flare FTSO delegation, P-chain staking (mirror), provider fee and validation rewards in FLR, the unclaimed epoch range, and Songbird FTSO rewards. Types owed but not yet priced on-chain…

  • get_my_portfolio_summary

    Return the calling wallet's FLR and SGB balances and their USD valuations. Native and wrapped (WFLR / WSGB) balances are reported separately; wrapped balances roll into the per-token USD values and the total. Requires OAuth scope…

  • get_my_rewards_history

    Return the calling wallet's reward history: FLR claimed per recent reward epoch (at least the last 30 days), split into FTSO delegation, P-chain staking and provider fee (including auto-claims), plus cumulative validation rewards and…

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "flarewatch": {
      "type": "http",
      "url": "https://mcp.flarewatch.io/api/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http flarewatch https://mcp.flarewatch.io/api/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://mcp.flarewatch.io/api/mcp as a custom connector in the settings.

Also shipped through: npm

OpenAPI spec

10 paths, read 2026-10-15. Open the spec ↗

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.