Web3 Systemic Agent
Web3 security, DEX honeypot detector, and DAO governance auditor for Base & EVM chains.
Description from the maker (original English)
Agent-safe fleet management for independent Solana validators and RPC nodes
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"solfleet": {
"command": "uvx",
"args": [
"solfleet",
"--from",
"solfleet",
"solfleet-mcp"
]
}
}
}Run this in your terminal
claude mcp add solfleet -- uvx solfleet --from solfleet solfleet-mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"solfleet": {
"command": "uvx",
"args": [
"solfleet",
"--from",
"solfleet",
"solfleet-mcp"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"solfleet": {
"type": "stdio",
"command": "uvx",
"args": [
"solfleet",
"--from",
"solfleet",
"solfleet-mcp"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Runs on your computer; the tool list appears in your AI app after you install it.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 9, 2026
The registry name was verified with the GitHub account sanjeevkkansal, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
This one runs on your computer; there is no internet endpoint
Runs on your computer and uses local settings instead
Runs on your computer; we can only read tool lists from internet endpoints
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).
The GitHub repo last changed –, and the latest version (0.1.1) reached the registry 4 months ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Python (uv) on your computer.
In the same category: Web3 Systemic Agent, Highwire, make.fun
Servers in the same category from other makers, sorted by stars.
Reading a card:↓ Downloads per weeki★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
Web3 security, DEX honeypot detector, and DAO governance auditor for Base & EVM chains.
Trade, launch and research higher.fun coins on Solana, from any AI agent. Unsigned or policy-signed.
Earn USDC on Solana by winning coding bounties: find a bounty, track entries, claim rewards.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Agent-safe fleet management for independent Solana validators and RPC nodes. One config file describes your fleet across devnet, testnet, and mainnet. An MCP server (and a CLI) exposes Solana-aware status, safe in-place upgrades, and health-driven DNS failover to Claude or any MCP client. Every operation that changes a node is dry-run by default, policy-gated, and audited. solfleet never reads or moves your keypairs.
See PLAN.md for the roadmap and design notes.
solfleet runs on the operator's machine (or a small VM). It talks to the fleet over JSON-RPC (read) and SSH/scp (act), builds artifacts on a separate build host, computes slot lag against each cluster's reference RPC, and manages failover records at the DNS provider. Every mutation flows through one gate and is written to a SQLite audit log.
flowchart TB
claude["Claude / any MCP client"]
subgraph operator["operator machine"] mcp["solfleet-mcp (stdio)"] cli["solfleet CLI"] core["core: probe · safety gate · executor · dns"] audit[("audit log (SQLite)")] claude -->|MCP| mcp mcp --> core cli --> core core --> audit end
builder["build host (agave + geyser from source)"] ref["cluster reference RPC"] dns["DNS provider (Cloudflare / Route53)"]
subgraph fleet["fleet: devnet / testnet / mainnet"] rpc["RPC nodes"] val["voting validators"] end
core -->|JSON-RPC :8899| rpc core -->|JSON-RPC :8899| val core -->|SSH / scp| rpc core -->|SSH / scp| val core -->|SSH build, fetch artifacts| builder builder -. "artifact set + sha256" .-> core core -->|slot lag / delinquency| ref core -->|eject / restore A records| dns
sequenceDiagram
actor Op as Claude / operator participant SF as solfleet participant B as build host participant N as node participant R as reference RPC Op->>SF: upgrade node to version (confirm) SF->>SF: gate, policy + preflight (else stop) SF->>B: build agave + geyser (or reuse cache) B-->>SF: artifact set + sha256 SF->>N: scp artifacts as dest.solfleet-new SF->>N: sha256 on node matches builder (else abort) alt RPC node SF->>N: systemctl stop SF->>N: atomic swap (binary + geyser + marker) SF->>N: systemctl start else voting validator SF->>N: atomic swap (binary + geyser + marker) SF->>N: agave-validator exit (leader-aware), systemd relaunches end loop until healthy and caught up SF->>R: getSlot SF->>N: getHealth / getSlot end SF->>SF: verify reported version, write audit entry
sequenceDiagram
participant SF as solfleet watch participant N as pool members participant R as reference RPC participant D as DNS provider loop every interval SF->>N: getHealth / getSlot SF->>R: getSlot (cluster head) SF->>SF: per member: unhealthy, lag over limit, or delinquent alt every member failing SF->>SF: keep current records (never empty the pool) else at least one healthy SF->>D: ensure TXT ownership marker SF->>D: remove A record of each failing member SF->>D: add A record of each recovered member SF->>SF: write audit entry end end
.so), caches it, and distributes the artifact set to the fleet.v1. Built and unit-tested (91 tests, CI on Python 3.11-3.13). Most paths are also proven live against a disposable devnet node and a real Cloudflare zone.
Proven live:
status, validate, vote-status, inspectrestart (RPC via systemctl; validator via leader-aware safe-exit)upgrade end to end (build agave from source on a builder, distribute, sha256-verify on the target, atomic swap, catch-up) for both RPC and voting-validator nodesbootstrap-builder (toolchain + deps on a bare builder)provision a voting validator from bare disks (format NVMe, install, render the voting unit, start, catch up, vote)dns status / eject / restore and last-member protection, against a live Cloudflare zoneUnit-tested but not yet run live:
watch loop (probe -> decide -> act); its decision logic is unit-tested and it reuses the now-proven Cloudflare driverNot built yet: HTTP transport (MCP is stdio-only today). See PLAN.md (M6).
pipx install solfleet # not yet published; for now:
pipx install git+https://github.com/sanjeevkkansal/solfleet pipx install 'solfleet[route53]' # if you use Route53 for DNS
cp fleet.example.yaml fleet.yaml # edit with your nodes
cp policy.example.yaml policy.yaml # optional; sane defaults if absent solfleet status # probe the fleet solfleet status --watch # refreshing live table solfleet validate # structural + live readiness check solfleet vote-status mn-val-1 # voting health: credits, balance, delinquency, leader solfleet inspect mn-val-1 # read-only SSH detail for one node solfleet bootstrap-builder b1 # install build toolchain on a builder; --confirm solfleet provision rpc-1 4.1.0 # dry-run bring-up plan; --confirm to run solfleet plan-upgrade mn-val-1 4.1.0 # dry-run upgrade plan solfleet upgrade mn-val-1 4.1.0 # dry-run; add --confirm to execute solfleet watch --dry-run # DNS failover loop, decide-only
MCP (Claude Code):
claude mcp add solfleet -- solfleet-mcp
Pointed at a small devnet fleet. With no flags, commands are read-only or dry-run.
Fleet health is Solana-aware, not just an HTTP 200:
$ solfleet status
CLUSTER NODE ROLE HEALTH VERSION SLOT LAG VOTE devnet rpc-1 rpc ok 4.1.0-rc.1 0 - devnet rpc-2 rpc ok 4.1.0-rc.1 0 -
An upgrade is dry-run by default. It returns the ordered plan and the gate decision and changes nothing until you pass --confirm:
$ solfleet plan-upgrade rpc-1 4.1.0
{ "decision": { "operation": "upgrade", "cluster": "devnet", "node": "rpc-1", "mode": "dry-run", "allowed": true, "plan": [ "on builder 'build-1': build agave 4.1.0 from source", "distribute artifact set to rpc-1; checksum-verify each (abort on mismatch)", "stop solana-validator, swap, start", "swap /usr/local/bin/agave-validator + geyser .so + version marker atomically", "wait until healthy + caught up to https://api.devnet.solana.com", "verify reported version == 4.1.0; record before/after" ], "reasons": [ "dry-run: preflight checks pass; pass confirm=true to execute" ] }, "target_version": "4.1.0" }
Over MCP, the same operations are tools (fleet_status, plan_node_upgrade, upgrade, ...). Claude gets that same plan back and has to pass confirm=true to execute, so an agent cannot mutate a node by accident.
Read-only: fleet_status, node_detail, version_drift, vote_status, leader_schedule, validate, plan_node_upgrade, dns_pool_status, audit_log.
Gated (dry-run by default; confirm=true to execute): bootstrap_builder_host, provision, restart, upgrade, dns_pool_eject, dns_pool_restore.
Every mutation is dry-run by default, checked against policy.yaml (allowed versions, disk floor, leader-window minimum), and written to a SQLite audit log. The watch loop is the one autonomous mutator; it is bounded by the same audit log and the never-empty-a-pool rule.
confirm=true.policy.yaml: allowed version globs, disk floor, and require_leader_window_minutes for validators.uv venv && uv pip install -e '.[dev]'
uv run pytest
Published to the MCP Registry.
mcp-name: io.github.sanjeevkkansal/solfleet