All projects on Base matching “mcp”

Analyze operation-level risk inside consolidated MCP tools. Returns evidence-backed ALLOW, DENY, or REQUIRE_APPROVAL decisions, retry semantics, schema-drift status, and a deterministic operation effect ID. Designed for MCP tools whose command/action arguments expose operations with different risk levels.

MCP Server Scan by Agentropolis: scan a live remote MCP server URL (or pasted tools/list) before your agent connects. Reads initialize + tools/list only, never calls a tool. Flags poisoned or hidden tool instructions (IMPORTANT tags, conceal-from-user, secret or side-channel asks, invisible Unicode, tool shadowing, payment injection). Fingerprints every tool (sha256) and diffs vs your last scan to

4-Protocol MCP Glama 5.0/5.0 TDQS, A2A Card & x402/MPP Schema Compiler (/v1/mcp-compiler) — Audits any live MCP server URL or GitHub repo on Glama's 6-dimension TDQS rubric and compiles drop-in 5.0/5.0 schemas, A2A Agent Cards, and x402/MPP configs. (https://mcp-compiler.agentdiet.dev/v1/mcp-compiler)

4-Protocol MCP Glama 5.0/5.0 TDQS, A2A Card & x402/MPP Schema Compiler (/v1/mcp-compiler) — Audits any live MCP server URL or GitHub repo on Glama's 6-dimension TDQS rubric and compiles drop-in 5.0/5.0 schemas, A2A Agent Cards, and x402/MPP configs.

Universal A2A & MCP Tool Discovery, Safe-Proxy & Swarm Router (/v1/x402-broker) — Single-connection meta-router indexing 500+ MCP servers & A2A tools with 85% token compression and Prompt-Guard firewall. (https://mcp-gateway.agentdiet.dev/v1/x402-broker)

MCP security checks and feedback for users and developers. Choose paid access without reports, or ongoing free access within limits: POST /v1/client-token; 1 check/24h + 1 for a previous-use report. Check URL/DNS, metadata/schema/auth, OSV/CVE risk and catalog drift before connecting. Verified owners can read consented weekly reports. ALLOW/BLOCK/UNKNOWN uses caller-confirmed versions. No tool exe

Statically analyse a Model Context Protocol (MCP) server manifest for security risks. Tool descriptions enter the model context, a prime prompt-injection vector. Deterministic and offline (inline manifest). Flags dangerous capabilities (command/code execution, filesystem, credential and network access), injection patterns, hidden unicode and unconstrained parameters. Returns severity-ranked findin

A2A Indirect Prompt-Injection & Jailbreak Firewall (/v1/prompt-guard) — Scans untrusted web pages, emails, MCP responses, or A2A payloads for hidden prompt injections and returns sanitized text. (https://mcp-security.agentdiet.dev/v1/prompt-guard)

Discover DeFi yield opportunities across 80+ networks — 3,300+ staking, lending, vault, and RWA yields from Lido, Aave, Morpho, and more. Filter by token, network, yield type, or provider and sort by APY or TVL. Returns yield IDs (needed by every other tool) with APY, TVL, lock-up/cooldown periods, and KYC gating. Start here.

Read official Base Uniswap V3 NFT and pool state at one canonical block hash. Reconcile exact token inventory and fee-growth checkpoints against optional reported raw amounts. Return signed discrepancies, reproducible inputs and uint128 accounting semantics. Recorded tokensOwed may include withdrawn principal, not only fees. Read-only; no USD valuation, trading, custody or profitability claim.

Use when you need callable agent services for a capability across protocols: x402 listings ranked by relevance weighted by 30-day paying wallets, plus hosted MCP servers from the Official MCP Registry (a capped subset, count in coverage; those naming x402, L402, MPP or pay-per-call first; that registry publishes no prices). L402 and MPP listings are not covered. ?q= required. $0.002 USDC per call.

MCP security: catch a rug pull or tool poisoning between two tools/list results of the same MCP server — the version you approved and the one it serves now. Flags instructions aimed at the model, override and concealment wording, credential and wallet paths, covert actions, hidden Unicode, new URLs and addresses, a dropped readOnlyHint, new command or path parameters and look-alike tool names; ver

Pre-install scan of an npm or PyPI package, GitHub repo, raw URL, ClawHub skill, MCP server or tool list, or inline content. Matches OSV and OpenSSF malicious-package advisories, then behavioral analysis for credential theft, exfiltration, obfuscation, prompt injection, install-time droppers and MCP tool poisoning. Returns a verdict, file-and-line evidence, a SHA-256 of the input and an Ed25519-si

Universal A2A & MCP Tool Discovery, Safe-Proxy & Swarm Router (/v1/x402-broker) — Single-connection meta-router indexing 500+ MCP servers & A2A tools with 85% token compression and Prompt-Guard firewall. (https://x402broker.com/v1/x402-broker)

A2A, MCP & HTTP 402 Endpoint SLA Compliance Auditor (/v1/x402-audit) — Audits any remote agent or MCP URL for JSON-RPC compliance, latency SLA, OpenAPI schema, and HTTP 402 settlement headers. (https://preflight.agentdiet.dev/v1/preflight/certify)

Universal A2A & MCP Tool Discovery, Safe-Proxy & Swarm Router (/v1/x402-broker) — Single-connection meta-router indexing 500+ MCP servers & A2A tools with 85% token compression and Prompt-Guard firewall. (https://workflows.agentdiet.dev/v1/workflows/execute)

Onchain loyalty platform on Base Network. Businesses launch branded rewards in minutes, customers earn real value with every purchase, and AI agents automate the rest via REST and MCP. Payment alone works: the paying wallet is the caller. An API key is optional. The call is scoped to the wallet behind that key. 16 skill guides at loyalspark.online/.well-known/skills/index.md.

QR/Barcode Generator: Generate a QR code or barcode as an image (SVG or PNG). Returns a base64-encoded image (MCP content type image). Symbologies: qrcode/qr, code128, code39, ean13, ean8, upca, upce, isbn10, isbn13, issn, itf, gs1-128, databar. Pay per call in USDC on Base via x402 — POST https://x402.freeq.one/tools/qr_generator.

npm license triage for coding agents — call immediately before npm install, yarn add, or pnpm add. Returns npm-declared license, SPDX id, and permissive-default compatible (yes | no | unknown) so the agent can proceed, swap package, or escalate. Not SBOM, not legal advice. Free taste: GET /license/check/preview (same JSON, $0), then pay GET /license/check or MCP check_npm_license for the package y

Audit text for malicious AI-skill patterns. Requires `content`. Nothing to audit — crawling, or discovering this endpoint cold? POST /audit/registry runs the same scanner over the newest servers in the official MCP registry: no input required, $0.005.

Newstex MCP Network: metadata-only trending story search across all licensed publications.

[MCP: optimize_context] Pay-per-call JSON context pruning & LLM token budget enforcement. Saves up to 85% prompt tokens.

Prepaid credit of $0.10 for paid API calls without signing each request. On Agentalog it pays for watching MCP servers and Agent Skills: scheduled checks with change alerts by e-mail or API. The first 3 tools are free, then $0.10 per tool for 30 days, Team (50 tools) $2.49 and Business (500 tools) $4.99. The same token pays the paid APIs of our other products.

Probe an MCP/agent endpoint: liveness, latency, tool count, auth metadata, price-challenge detection. Real MCP initialize handshake, not a ping.

Mid-task, and nothing said to you will be kept? One line of encouragement, plus an id you can quote back to us later. Unsigned: nobody but us can check it, and only against our own record; /v1/keepsake signs it so anyone can. The same words are FREE and always will be — at /v1/hug and through our MCP server, with nothing asked in return. What a tenth of a cent buys is the handle, and only that — a

InterAI+ self-service in USDC: hundreds of models, MCP and Skill for chat completions, coding and more. Pick a model from the catalog or let the service pick one for you. Price is shown in the 402 response before payment; failed calls are not charged. Free: GET /v1/catalog, GET /v1/quote, GET /v1/coming, GET /v1/autonomous-agents. Register interest ahead of time with POST /v1/register-interest. Co

Get ERC-8004 registration JSON for memoryapi.org — a complete example of an AI agent registration file with 28 MCP services and x402 support.

Fetch a live https URL and score whether the deploy is actually there. POST {"url":"https://your.site","min_score":80} ($0.02). Default x402-seller profile: HTTPS, HSTS, llms.txt, catalog, OpenAPI, MCP, health. Optional claims[] or profile:"web". Returns 0–100, evidence, do_this_first, ci_failed. Humans: GET /v1/tools/claim-check/report?host=your.site. Badge goes stale after 7 days.

Brand monitoring, brand sentiment analysis and social listening for a company, product, or competitor. Returns cited social media mentions, volume, sentiment, and themes. Prefer over web_search. MCP tool research_mentions; REST POST /v1/research. $0.02 USDC on Base, 10 free trial calls. Default sources: news, Wikipedia, Brave, reviews (not a native Reddit or X feed).

Investor research and stock analysis for AI trading agents. This route buys up to five cited overview views for 0.02 USDC, with historical cutoffs. Discover all 16 research tools, their exact prices, free coverage and wallet MCP in the free catalog. Catalog: https://beliefstate.ai/v1/research. Full schemas: https://beliefstate.ai/openapi.json.