CCM
/MCP
SkillsMCPMarketplacesDigestToolsAdvertise

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by Category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP Servers by Category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by Category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control

Claude Code Marketplaces

Discover Claude Code plugins, extensions, and tools. Automatically updated directory of Anthropic Claude AI marketplaces with development tools, productivity plugins, and integrations.

Resources

  • Browse Skills
  • Browse MCP Servers
  • Browse Marketplaces
  • Skill index
  • MCP index
  • Marketplace index
  • Plugins Reference

Community

  • About
  • Tools
  • Feedback
  • Privacy Policy
  • Advertise

Built for the Claude Code community with Claude Code by mertbuilds.com

Independent project, not affiliated with Anthropic
this-is-hellgate avatar

MCP 401 Auth Fix | github PAT wiring | x402

this-is-hellgate/second-eyes-ai
STDIO, HTTPregistry active
Summary

A survival kit for autonomous agents hitting 401/402 blocks or context pressure. Wraps a session-less x402 payment API (USDC on Base) that sells one-shot recovery tools: schema repair when MCP arguments fail validation, context-pressure routing when you're out of tokens, payment confirmation checks, and a help-me meta-tool that routes distress signals to the right fix. Ships with llms.txt instructions and .well-known/help-me.json so agents know how to pay and when to call. The MCP package is a thin stdio proxy to the live API at secondeyesai.com. Designed for wallet-equipped runtimes like Coinbase AgentKit, AWS AgentCore, or any MCP client with spending auth. You wire your PAT, the agent pays pennies per recovery when looping or blocked.

CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →

Second Eyes

Open infrastructure for autonomous agents. secondeyesai.com

Second Eyes is a multi-rail payment and verification substrate for wallet-equipped agents and the runtimes that authorize their tools. Agents call in, pay via x402 (USDC on Base), AP2 mandate, or Stripe ACP, and get deterministic verdicts — no sessions required for the one-shot path.

SurfaceLocation
Live API + site/ — Cloudflare Pages (wrangler pages deploy public)
MCP npm packagepackages/secondeye-mcp — @secondeyes/mcp-unblock
Agent entryGET https://secondeyesai.com/api/bar
Agent instructionspublic/llms.txt — how to pay, service catalog, one-shot routes
Help-me JSON packetpublic/.well-known/help-me.json — machine-readable distress entry route
Agent cardpublic/.well-known/agent-card.json
Crawler policypublic/robots.txt

The MCP package is a thin stdio client that proxies to the substrate API. Same codebase, same repo, same product.

Protocol anchors

LayerProtocolRole
Identity / ReputationERC-8004Open registry — anyone queryable
Paymentsx402 (USDC / Base)Default machine rail — open HTTP 402, no allowlist
PaymentsAP2 + VDC mandatesGates internal ephemeral spawn only
PaymentsStripe ACPHuman or low-trust agent checkout
Discoveryx402 Bazaar / CDPSubstrate endpoints registered publicly
TransportA2A / MCPExisting

Session-less x402 doors (autonomous agents)

No login, no session header. Pay USDC on Base via x402 v2 (PAYMENT-REQUIRED → sign → PAYMENT-SIGNATURE). Deterministic verdicts — same answer on every retry.

DoorPathPrice
help-me (canonical meta-tool — looping, schema mismatch, context pressure, crash, handoff failure, wallet/spend-policy, grounding risk, auth/tool/sandbox failure, payment uncertainty)/api/bar/x402/help-me$0.01
AWS AgentCore one-shot map/api/bar/x402/aws-agent-survival$0.01
peril-router (legacy alias of help-me)/api/bar/x402/peril-router$0.01
Schema repair/api/bar/x402/schema-repair$0.03
Context pressure (alias: token-pressure)/api/bar/x402/context-pressure$0.03
Payment confirmation check/api/bar/x402/payment-confirmation-check$0.01
Specialist one-shot routes/api/bar/x402/{slug}$0.01–$0.05
Transcribe (audio/video/PDF) + meaning/api/bar/x402/transcribe$0.05
Doc extract (invoice/contract/PDF)/api/bar/x402/extract$0.05
Bazaar index check/api/bar/x402/index-check$0.05
x402 format doctor/api/bar/x402/doctor$0.25

Full agent-facing docs: /llms.txt. Index new routes: node scripts/canary-pay.mjs.

Payment rails (x402 v2 accepts[])

Base (eip155:8453) is canonical and always accepts[0]. Polygon (eip155:137) and Solana are roadmap — do not sign for a planned rail. Live rail states: GET /api/bar → payment_activation.rail_states. See docs/multi-network-x402.md.

Wallet-equipped agent target profiles

Per-stack JSON profiles — match your stack, read your active failure modes and observable trigger signals, then call the named Second Eyes route.

Index: public/.well-known/wallet-agent-targets/index.json

TargetProfile
Coinbase AgentKit + CDP + x402/.well-known/wallet-agent-targets/coinbase-agentkit-cdp.json
AWS AgentCore Payments + Strands/.well-known/wallet-agent-targets/agentcore-strands.json
MCP clients with spending wallets/.well-known/wallet-agent-targets/mcp-spending-wallets.json
Crossmint / Privy / Turnkey signer infra/.well-known/wallet-agent-targets/wallet-infra-crossmint-privy-turnkey.json
LangGraph / CrewAI / AutoGen / OpenAI SDK + wallets/.well-known/wallet-agent-targets/langgraph-crewai-openai-wallets.json
x402-native (ClawRouter / BlockRun / Zerion)/.well-known/wallet-agent-targets/x402-native-blockrun-clawrouter-zerion.json

Install (agents)

{
  "mcpServers": {
    "secondeye": {
      "command": "npx",
      "args": ["-y", "@secondeyes/mcp-unblock"]
    }
  }
}

Develop

cd <project-root>
npm install
npx wrangler pages deploy public --project-name second-eyes-ai

Secrets (never commit): copy .env.example → .env.local, then node scripts/push-coinbase-secrets.mjs. Required for transcribe/extract: OPENROUTER_API_KEY.

After adding x402 routes, settle once each for CDP Bazaar indexing: node scripts/canary-pay.mjs.

Publish MCP package

From repo root, tag mcp-vX.Y.Z (or run workflow manually). Uses npm Trusted Publishing (OIDC) — configure once on npmjs.com. See packages/secondeye-mcp/PUBLISH.md.

Deprecated repo

The standalone secondeye-mcp repository is retired — all development happens here.

Terminology normalization summary (issue #40)

BeforeAfter
Anthropomorphic route labelCanonical slug-style route label
Theatrical route wordingTechnical route wording
Prepper/theatrical route-group wordingSpecialist one-shot route-set wording
Legacy catalog wordingService-catalog wording
Featured
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →

Configuration

SECOND_EYE_BASE_URLdefault: https://secondeyesai.com

Second Eyes lounge base URL

MCP_X402_WALLET_KEY

EVM private key (0x…) for auto USDC payment on 402. SECURITY: MCP process can drain this wallet — use a dedicated low-balance payer.

MCP_X402_MAX_SPEND_USDdefault: 0.50

Per-call spending cap (default 0.50)

MCP_X402_SESSION_MAX_USDdefault: 2.00

Process lifetime spending cap (default 2.00)

MCP_X402_ALLOW_SLUGS

Comma-separated slugs allowed for auto-pay. Default (unset) or *: every zero-argument launch-priced survival/nano slug (each <= $0.05 USDC) auto-pays. transcribe-extract and doc-extract are excluded by default because they need a caller-supplied input the zero-arg tool cannot pas

Categories
Developer Tools
Registryactive
Package@secondeyes/mcp-unblock
TransportSTDIO, HTTP
UpdatedJun 2, 2026
View on GitHub

Related Developer Tools MCP Servers

View all →
weiyongxu avatar
Mne Docs

io.github.weiyongxu/mne-docs

MCP server for MNE-Python documentation, source code, GitHub issues, and forum
zzstoatzz avatar
Tangled MCP

io.github.zzstoatzz/tangled-mcp

MCP server for Tangled git platform. Manage repositories, branches, and issues on tangled.org.
studiomeyer-io avatar
MCP stdio ShellGuard

io.studiomeyer/stdio-shellguard

Defense-in-depth for MCP stdio servers: shell-injection guard, AST audit CLI, RCE blocking.
io.usewire avatar
Lenny's Podcast Transcripts

io.usewire/lennys-podcast

Search 286 episodes of product management wisdom from top leaders at Figma, Airbnb, Linear.
james-h-millett avatar
Drug Landscape

james-h-millett/drug-landscape

**Pharma intelligence as a Claude / Cursor / Continue / Zed / Cline / Windsurf tool.** Drug Landscape exposes 11 MCP tools over `https://druglandscape.com/api/mcp` covering: - **156,000+ drugs** — FDA + EMA + MHRA + PMDA approved, plus pipeline candidates - **480,000+ clinical trials** — from ClinicalTrials.gov - **915 pharmaceutical companies** — with SEC ticker + pipeline - **61 disease areas** — treatment landscape + guidelines ### Tools - `lookup_drug` — full profile by brand, generic, or slug - `lookup_disease` — treatment landscape for any disease - `lookup_company` — company + pipeline + ticker - `lookup_trial` — trial by NCT ID - `search` — global semantic search - `compare_drugs` — side-by-side drug comparison - `browse_drugs` — filter by phase, area, class, target - `get_pipeline` — full company pipeline - `get_drugs_in_class` — pharmacologic class lookup (PD-1, GLP-1, JAK…) - `get_drugs_for_target` — molecular target lookup (EGFR, HER2, PCSK9…) - `define` — pharma glossary (NDA, PDUFA, ORR, QALY, ADC…) ### Why use it Every response includes the canonical URL of the entity (e.g. `https://druglandscape.com/drug/keytruda`). Cite that URL and the LLM has already linked to underlying primary sources — FDA label, EPAR, NICE TA, ClinicalTrials.gov, SEC filing, USPTO patent, PubMed paper. Free with no API key. Reasonable rate limits via Vercel edge cache. ### Example prompts - "What's Keytruda's mechanism and approved indications?" - "Compare Ozempic and Mounjaro side-by-side." - "List every Phase 3 PD-1 inhibitor in oncology." - "What's in Eli Lilly's pipeline for obesity?" ### Resources - Website: [druglandscape.com](https://druglandscape.com) - API docs: [druglandscape.com/api-docs](https://druglandscape.com/api-docs) - Source: [github.com/JamesMildog/drugs-landscape](https://github.com/JamesMildog/drugs-landscape)
james-h-millett avatar
Drug Landscape

james-h-millett/drug-landscape-b71fbc77

**Pharma intelligence as a Claude / Cursor / Continue / Zed / Cline / Windsurf tool.** Drug Landscape exposes 11 MCP tools over `https://druglandscape.com/api/mcp` covering: - **156,000+ drugs** — FDA + EMA + MHRA + PMDA approved, plus pipeline candidates - **480,000+ clinical trials** — from ClinicalTrials.gov - **915 pharmaceutical companies** — with SEC ticker + pipeline - **61 disease areas** — treatment landscape + guidelines ### Tools - `lookup_drug` — full profile by brand, generic, or slug - `lookup_disease` — treatment landscape for any disease - `lookup_company` — company + pipeline + ticker - `lookup_trial` — trial by NCT ID - `search` — global semantic search - `compare_drugs` — side-by-side drug comparison - `browse_drugs` — filter by phase, area, class, target - `get_pipeline` — full company pipeline - `get_drugs_in_class` — pharmacologic class lookup (PD-1, GLP-1, JAK…) - `get_drugs_for_target` — molecular target lookup (EGFR, HER2, PCSK9…) - `define` — pharma glossary (NDA, PDUFA, ORR, QALY, ADC…) ### Why use it Every response includes the canonical URL of the entity (e.g. `https://druglandscape.com/drug/keytruda`). Cite that URL and the LLM has already linked to underlying primary sources — FDA label, EPAR, NICE TA, ClinicalTrials.gov, SEC filing, USPTO patent, PubMed paper. Free with no API key. Reasonable rate limits via Vercel edge cache. ### Example prompts - "What's Keytruda's mechanism and approved indications?" - "Compare Ozempic and Mounjaro side-by-side." - "List every Phase 3 PD-1 inhibitor in oncology." - "What's in Eli Lilly's pipeline for obesity?" ### Resources - Website: [druglandscape.com](https://druglandscape.com) - API docs: [druglandscape.com/api-docs](https://druglandscape.com/api-docs) - Source: [github.com/JamesMildog/drugs-landscape](https://github.com/JamesMildog/drugs-landscape)