Coding Agent Package Hallucination & Slopsquatting Guard MCP
Server Details
Verify npm & PyPI package existence and age before install to block slopsquatting attacks.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-03-26
- URL
- Repository
- BeakDo/defi-customs-mcp-hub
- GitHub Stars
- 0
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 5 tools
The four crypto tools each target a distinct mechanism (external telemetry audit, permission struct generation, swap calldata assembly, local tx simulation), but plan_and_build_guarded_rebalance_swap and simulate_evm_tx_and_spend_firewall both enforce spend caps over swap transactions, creating real boundary ambiguity about which to call. audit_token_contract_technical_risks also partially overlaps the firewall's dangerous-approval detection, just via a different data source. An agent could plausibly misselect between the swap-builder and the simulator.
All five names are snake_case and verb-first (audit_, build_, plan_and_build_, simulate_, verify_), which is a clear and predictable pattern. The deviation is stylistic bloat: multi-clause names using 'and' and protocol numbers (build_erc7715_and_coinbase_spend_permission, plan_and_build_guarded_rebalance_swap) make lengths wildly uneven, but the convention itself holds.
Five tools is on the lean side but each maps to a real step in a guarded agent-transaction workflow (vet contract, grant bounded permission, assemble swap, simulate + budget firewall, verify dependency). Nothing looks like filler, though the surface is thin enough that a few more operations would be reasonable.
The server's title promises package hallucination/slopsquatting guarding, yet only one tool addresses it and it only checks existence plus registration age — no typosquat similarity, license/metadata checks, transitive dependency or lockfile auditing, or suggested safe alternatives. The crypto side is also partial: no approval revocation, no executed transfer tool, and no standalone balance/quote retrieval, so agents hit dead ends on remediation paths.
Available Tools
5 toolsaudit_token_contract_technical_risksAudit ERC-20 Token Bytecode for Honeypot, Tax & Liquidity RisksARead-onlyIdempotentInspect
Checks GoPlus and DexScreener telemetry for honeypot traps, buy/sell tax %, hidden owner privileges, and live pool liquidity before swapping.
Inputs: chainId, tokenAddress.
build_erc7715_and_coinbase_spend_permissionBuild ERC-7715 Session Key & Coinbase SpendPermission PayloadsARead-onlyIdempotentInspect
Generates non-custodial ERC-7715 wallet_grantPermissions JSON-RPC and Coinbase SpendPermissionManager EIP-712 structs to bound AI agent daily/per-tx USDC spending.
Inputs: chainId, maxPerTxUsdc, smartAccount, agentSessionSigner, +1 more.
plan_and_build_guarded_rebalance_swapPlan Portfolio Drift & Build Spend-Capped 0.25% Multi-DEX Swap CalldataBRead-onlyIdempotentInspect
Reads live ETH + USDC wallet balances, enforces maxSpendCapUsd, and assembles ready-to-sign non-custodial multi-DEX swap calldata saving 71% vs standard wallet swap fees.
Inputs: chainId, targetEthPct, walletAddress, maxSpendCapUsd.
simulate_evm_tx_and_spend_firewallPre-Flight Simulate EVM Transaction for Reverts, Gas & Spend CapARead-onlyIdempotentInspect
Simulates any unsigned EVM transaction via eth_call + eth_estimateGas, detects dangerous unlimited ERC-20 approvals, and verifies total USD outflow against the agent's budget cap.
Inputs: to, data, from, chainId, +2 more.
verify_package_anti_hallucinationVerify npm or PyPI Package Against AI Hallucination & SlopsquattingARead-onlyIdempotentInspect
Checks live npm and PyPI registries to block nonexistent hallucinated package names and flag newly registered slopsquatting traps (<14 days old).
Inputs: ecosystem, packageName.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Check npm and PyPI packages exist before install: versions, age, downloads, look-alikes.
Blocks typosquatted or hallucinated npm/PyPI packages before an AI agent installs them.
Check npm packages for typosquats, hallucinated names, install scripts and risky new releases.
Verify npm packages before your AI agent installs them: hallucinations, advisories, API drift.
Related MCP Servers
- AlicenseAqualityBmaintenanceVerifies npm packages for security risks before installation, checking advisories, install scripts, typosquatting, and other factors, providing safe/block verdicts.160 npmApache 2.0
- AlicenseAqualityCmaintenancePre-install risk gate for npm packages. Stops AI coding agents from running malicious or lifecycle scripts run.12MIT
- AlicenseBqualityBmaintenancePre-install gate for AI coding agents. Checks npm, PyPI and crates.io package names before an install runs and blocks names that do not exist, were registered after models invented them, or sit one keystroke from a popular package. CLI, GitHub Action and MCP server. Apache 2.0.1306 PyPI1Apache 2.0
- AlicenseNot gradedqualityCmaintenanceEnables AI coding agents to automatically verify npm packages against the live registry before installation, flagging hallucinated, slopsquatted, or otherwise suspicious packages with risk verdicts.4 npmMIT
Glama MCP Gateway
Add one secure layer between your agents and this server.