{"success":true,"m2mEngine":"Panthera Proprietary M2M Cloud v2.2","strategyId":"GRD-M2M-SOCIAL-VIDEO-BOOM","strategyTitle":"\"One Call and Boom\" Social Video & Reel Unlocker (/api/v1/video/jobs?wait=true)","version":2,"settledPriceUsdc":0,"freeTrialSampleUsed":true,"freeCallsUsedToday":1,"freeCallsLimitPer24h":3,"freeCallsRemaining":2,"nextCallRequiresPaymentUsdc":0,"recipientTreasury":"CLBg2XGPZZ5h9tfUX1aAvLWtczbTuojJ4RNnHPrKyk1Y","targetInspected":"solana mcp agent","simulatedInAppTest":false,"http402ChallengeSpec":{"x402Version":2,"scheme":"exact","network":"solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp","asset":"EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v","assetSymbol":"USDC","maxAmountRequired":"1000","priceUsdc":0.001,"catalogPriceUsdc":0.04,"autoPilotDiscountApplied":true,"payTo":"CLBg2XGPZZ5h9tfUX1aAvLWtczbTuojJ4RNnHPrKyk1Y","resource":"https://rklhfyeyw0.execute-api.ap-south-1.amazonaws.com/api/x402/m2m/m2m-social-video-reel-boom","description":"\"One Call and Boom\" Social Video & Reel Unlocker (/api/v1/video/jobs?wait=true) ($0.001 USDC per call)","mimeType":"application/json"},"agentReach":{"schema_version":1,"ok":true,"channel":"web","action":"search","backend":"exa-mcp","items":[{"title":"solana mcp agent","url":null,"author":null,"published_at":null,"text":"Title: Solana Developer MCP\nURL: https://mcp.solana.com/\nPublished: N/A\nAuthor: N/A\nHighlights:\nSolana Developer MCP\n\n# Current Solana docs and Rust checks for coding agents\n\n Connect your MCP client to live Solana documentation retrieval, semantic search, and the Program autofixer for Anchor and Pinocchio program checks. \n\nReady\n\n Endpoint https://mcp.solana.com/mcp Transport Streamable HTTP Registered tools 5 Program autofixer Anchor + Pinocchio \n\n## Setup\n\nUse the direct HTTP endpoint when your client supports it. No API key is required.\n\n1\n\n### Add the server\n\nRun this in a terminal where the Codex CLI is installed:\n\n```\ncodex mcp add solana-mcp --url https://mcp.solana.com/mcp\n```\n\n2\n\n### Verify the connection\n\nList configured MCP servers and confirm `solana-mcp` appears:\n\n```\ncodex mcp list\n```\n\n3\n\n### Add project instructions\n\n Put the agent instructions below in `AGENTS.md` when you want Codex to route Solana work through this MCP server consistently. \n\n### Add the server\n\nClaude Code supports remote HTTP MCP servers directly:\n\n```\nclaude mcp add --transport http solana-mcp https://mcp.solana.com/mcp\n```\n\n### Verify the connection\n\nOpen Claude Code and run `/mcp`. The server should list these tools:\n\n```\nlist_sections\nget_documentation\nSolana_Documentation_Search\nSolana_Expert__Ask_For_Help\nprogram_autofixer\n```\n\n### Configure MCP\n\n Add this to `~/.cursor/mcp.json` for global use, or `.cursor/mcp.json` inside one project: \n\n```\n{\n  \"mcpServers\": {\n    \"solanaMcp\": {\n      \"url\": \"https://mcp.solana.com/mcp\"\n    }\n  }\n}\n```\n\n### Verify in Agent chat\n\n Open Cursor Agent, check the available MCP tools, and ask it to call `list_sections` for Solana documentation discovery. \n\n### Configure MCP\n\nAdd this to `~/.codeium/windsurf/mcp_config.json`:\n\n```\n{\n  \"mcpServers\": {\n    \"solanaMcp\": {\n      \"serverUrl\": \"https://mcp.solana.com/mcp\"\n    }\n  }\n}\n```\n\n...\n\n### Verify in Cascade\n\n Open Cascade MCP settings, confirm `solanaMcp` is connected, and enable the Solana tools you want available.\n\n...\n\n### Open user configuration\n\nIn the Command Palette, run `MCP: Open User Configuration`, then add:\n\n```\n{\n  \"servers\": {\n    \"solanaMcp\": {\n      \"type\": \"http\",\n      \"url\": \"https://mcp.solana.com/mcp\"\n    }\n  }\n}\n```\n\n### Start and verify\n\n Run `MCP: List Servers`, start `solanaMcp`, then use Copilot Agent mode to ask a Solana question that requires current docs. \n\n## Tools\n\nThe live server currently exposes these exact MCP tool names.\n\n### `list_sections`\n\nLists indexed Solana documentation sources with section tags and use cases.\n\n### `get_documentation`\n\nFetches canonical docs by source id, section id, or an array of ids.\n\n### `Solana_Documentation_Search`\n\nRuns semantic search over the Solana docs corpus for narrow questions.\n\n### `Solana_Expert__Ask_For_Help`\n\nAnswers Solana how-to and debugging questions with retrieved docs context.\n\n### `program_autofixer` New\n\n Analyzes Solana program Rust for Anchor and Pinocchio security antipatterns and returns structured issues plus fix suggestions.\n\n...\n\n## Agent Instructions\n\n Add the matching file for your agent. Cursor project rules use `.cursor/rules/*.mdc`; `.cursorrules` is legacy. \n\n Add this to `AGENTS.md`. The key rule is the Program autofixer loop: call it before returning Solana program Rust, apply fixes, and call it again until another pass is not required. \n\n```\nFor Solana-related work, prefer the Solana Developer MCP tools over model memory.\n\nUse `list_sections` first for non-trivial Solana questions so you can find the\nright documentation source ids and section ids.\n\nUse `get_documentation` when you need canonical docs for a specific source,\nframework, library, or ecosystem area. Use `Solana_Documentation_Search` or\n`Solana_Expert__Ask_For_Help` for narrow how-to questions, errors, or API usage.\n\nWhenever you write or modify Solana program Rust, call `program_autofixer` before\nreturning code. It accepts `code`, optional `filename`, and optional `framework`\n(`auto`, `anchor`, or `pinocchio`). Apply the suggested fixes, then call\n`program_autofixer` again. Repeat until `require_another_tool_call_after_fixing`\nis false.\n```\n\n...\n\n### Canonical docs\n\nUse `list_sections`, then fetch the current Anchor docs for account constraints.\n\n### Semantic search\n\nSearch the docs for Token-2022 transfer hook account requirements.\n\n### Solana debugging\n\nAsk for help with a failed transaction log and include the program error output.\n\n### Rust review\n\nRun `program_autofixer` on this Pinocchio instruction before returning the patch.\n\n...\n\nSolana Developer MCP is served at `https://mcp.solana.com/mcp`.\n\n---\n\nTitle: Coding with agents\nURL: https://solana.com/docs/intro/coding-with-agents\nPublished: N/A\nAuthor: N/A\nHighlights:\n## Using Solana MCP\n\nThe most powerful way to leverage AI for Solana development is through the\nSolana Developer MCP (Model Context Protocol). This\nintegration brings specialized Solana knowledge directly into your IDE, giving\nyour agents access to up-to-date Solana documentation and making development\nfaster and more efficient.\n\n## What is Solana MCP?\n\nSolana MCP is a specialized AI assistant that integrates directly into\nAI-supported IDEs like Cursor and Windsurf. When you ask Solana-specific\nquestions, it automatically queries the MCP server to provide accurate,\nup-to-date information from:\n\n- The Solana documentation\n- The Anchor Framework documentation\n- Solana\nProgram Examples\n- Solana Stack Exchange questions and\nanswers\n\n## Example use cases\n\nYou can use MCP to:\n\n- Ask about Solana concepts\n- Debug program errors\n- Get best-practice recommendations\n- Learn from expert examples\n- Develop faster with the help of an expert\n\n## Set up Solana MCP\n\nFind a detailed setup and getting started guide on the\nSolana MCP website.\n\n## Prefer to use a skill?\n\nIf you'd rather use a skill, install the official Solana skill with:\n\n```bash\nnpx skills add https://github.com/solana-foundation/solana-dev-skill\n\n```\n\nBrowse more reusable workflows in the Solana Agent Skills library.\n\n---\n\nTitle: AI on Solana | Agents, Payments, and Compute\nURL: https://solana.com/solutions/ai\nPublished: N/A\nAuthor: N/A\nHighlights:\n# Build AI agents on Solana.\n\nGive agents verifiable identity and a way to pay for APIs, data, and compute. Use open tools to build, register, and run agents on Solana.\n\n## What Solana adds to AI\n\nAI models train and run offchain. Solana provides a shared layer for payments and coordination: agents can register identities, pay for digital services in stablecoins, and use programs to enforce spending rules. Teams building compute, training, and data networks use the same rails to coordinate contributors.\n\n...\n\n## AI teams building on Solana\n\nInference.net serves AI models through an API and coordinates distributed GPU capacity for inference jobs. Learn More\n\nNous Research's Psyche network coordinates distributed model training on Solana, tracking participants and training runs. Learn More\n\nGrass connects contributors who collect web data with teams that use data to train AI models. Learn More\n\nGradient develops distributed systems for model inference and reinforcement learning. Learn More\n\nelizaOS provides an open-source framework for building agents, with plugins that connect them to Solana. Learn More\n\n...\n\n## Build with AI on Solana\n\nChoose an entry point for coding, agent identity, or machine payments.\n\n---\n\nTitle: solana-foundation/solana-mcp\nURL: https://github.com/solana-foundation/solana-mcp-official\nPublished: N/A\nAuthor: N/A\nHighlights:\n# Official Solana MCP Server\n\nTry it out at https://mcp.solana.com !\n\nThe official Solana Developer MCP. Purpose: serve up-to-date documentation across the Solana ecosystem to AI agents and developer tooling.\n\n## Architecture\n\n- **Ingestion** (`ingestion/`): Databricks notebook crawls the sources listed in `ingestion/sources.yaml`, chunks markdown, and writes embeddings into a Delta-backed Vector Search index.\n- **Retrieval** (`lib/services/databricks/`): MCP tools query the index via the Databricks managed AI Search MCP server, which retrieves and reranks results server-side. `get_documentation` falls back to a SQL read of the `docs_chunks` Delta table when a source has no published `llms.txt`.\n- **Server** (`lib/index.ts`, `server/cloudrun.ts`): Exposes five tools over MCP — `Solana_Expert__Ask_For_Help` and `Solana_Documentation_Search` (semantic RAG), `list_sections` and `get_documentation` (canonical-spec retrieval modelled after the Svelte AI server), and `program_autofixer` for Anchor and Pinocchio Solana program Rust checks. Deployed on Cloud Run as a containerised Node service fronting `mcp.solana.com`; calls the Databricks workspace REST API directly for retrieval.\n- **Section catalogue** (`ingestion/sources.yaml` → `lib/sources.generated.ts`): `pnpm gen:sources` emits a typed catalogue of every source, its tags from a closed 21-section taxonomy, and `use_cases` keywords used by `list_sections` to route the agent.\n- **Analytics** (`lib/services/s3/analytics.ts`): Tool calls + initializations are buffered in memory and uploaded as JSONL objects to the S3 prefix configured by `ANALYTICS_S3_URI`.\n\n## CLI\n\n`cli/` ships `@solana/mcp`, a dependency-free command-line client for agents and scripts without MCP support. Each command is one `tools/call` against `mcp.solana.com`, so it needs no credentials.\n\n```bash\nnpx @solana/mcp search \"derive a PDA with Anchor\"\nnpx @solana/mcp ask \"why does my CPI fail with PrivilegeEscalation?\"\nnpx @solana/mcp sections\nnpx @solana/mcp docs anchor-docs frameworks\nnpx @solana/mcp check programs/vault/src/lib.rs   # or: cat lib.rs | npx @solana/mcp check -\n```\n\nAdd `--json` for the raw tool result, and `--framework anchor|pinocchio` to `check` when the framework is not detected from imports. Exit codes: `0` ok, `1` `check` found a syntax error or an undismissed critical/high issue, `2` usage, network, or server error. Set `SOLANA_MCP_URL` to point at another server, e.g. `http://localhost:8080/mcp`.\n\nTo release, bump `version` in `cli/package.json`, merge, then run the `Publish CLI` workflow from `main` (dry run first). It publishes to npm through Trusted Publishing and creates a `cli-v ` GitHub release.\n\n---\n\nTitle: Helius for Agents: Solana Developer Tools\nURL: https://www.helius.dev/agents\nPublished: N/A\nAuthor: N/A\nHighlights:\n# Helius for Agents: Solana Developer Tools\n\n> The complete toolkit for agents building on Solana — Helius Claude Code Plugin, MCP, Skills, CLI, and optimized SDKs.\n\nHelius for Agents\n\n## One-shot anything on Solana\n\nThe complete toolkit for agents building on Solana. One-shot Solana apps using the official Helius Claude Code Plugin, MCP Server, skills, and CLI.\n\nFor humans | For agents\n\n## The most comprehensive AI toolkit for building apps on Solana\n\n## Helius Claude Code Plugin\n\nUse every Helius API and tool like an expert directly from Claude Code. Query RPCs, parse transactions, and send transactions without leaving your editor.\n\n- Native RPC, WSS, DAS, Sender, and more\n- Secure API key handling and best practices\n- No context switching; build and ship in one flow\n\n## Helius MCP\n\nAccess 60+ Solana operations exposing RPC, DAS, webhooks, and transaction sending to any MCP-compatible AI assistant or agent framework.\n\n- Drop-in MCP server for Claude, Codex, and more\n- 60+ operations for balances, tokens, and transactions\n- Configure once, use across all agent workflows\n\n...\n\n## Helius CLI\n\nInstall the Helius CLI to authenticate and use 95+ Helius commands directly from your terminal. Ideal for scripting, CI/CD, and local agent tooling.\n\n- Works with all Helius products\n- 95+ commands from auth, to reads and writes\n- Use in scripts and automation via env or config\n\n...\n\n## Helius SDKs\n\nThe Official Rust and TypeScript SDKs for Helius, fully optimized and documented for agentic development.\n\n- TypeScript and Rust with full type definitions\n- Fully documented patterns for agentic workflows\n- RPC, DAS, WebSockets, and transaction helpers\n\n...\n\n### What can agents do with Helius?\n\nHelius gives AI agents direct access to Solana via the fastest, most reliable infrastructure possible. Query wallet balances, parse and send transactions, monitor addresses via webhooks, stream real-time data, estimate priority fees—the options are endless. Agents can create a Helius account and configure an API key entirely on their own, with no browser needed.\n\n### How do the Helius agent tools work?\n\nHelius offers a suite of products that give AI the knowledge and live access it needs to build on Solana. The MCP Server includes 60+ tools that AI agents call directly against Solana. Structured inputs, structured outputs, no shell commands or parsing. It covers all available Helius products and includes autonomous account signup. Works with Claude Code, Cursor, Codex, and any MCP-compatible tool. Skills are instruction sets that make Claude a Solana specialist. Each skill routes your request to the right tools and reference files, follows correct SDK patterns, and avoids common mistakes. Available for general Solana development with Helius (Build), trading apps (via DFlow), frontend apps (via Phantom), and protocol internals (SVM). The Claude Code Plugin bundles the MCP server with expert skills. Install to get live Solana access with deeply informed coding guidance all in one step. The Helius CLI provides shell-level access to the same capabilities as the MCP server. Useful for CI/CD pipelines, scripting, and environments where MCP isn’t available. The TypeScript and Rust SDKs provide type-safe wrappers around Helius APIs for agents so they can call Helius directly from code, without going through an AI tool.\n\n...\n\n### Which tool should I choose?\n\nWe recommend using the MCP for most use cases—it works with Claude Code, Cursor, Codex, and any MCP-compatible tool. Use the Claude Code plugin if you’re an avid Claude user and want skills bundled in with the MCP. Use the CLI for shell scripts and CI/CD pipelines where you need terminal-level automation. Use the TypeScript or Rust SDKs to write your own agent and have type-safe API access directly in code.\n\n...\n\n### Do I need a Helius account before I start?\n\nNo. An AI agent can programmatically create a Helius account for you with your email and name. The agent generates a Solana wallet and starts signup, you pay the $10 USDC one-time Agent plan fee from any wallet via the payment link (or the agent autopays from its funded wallet), and you get an API key and 1,000,000 credits. Existing users can set their API key and use their plan to the fullest across any of our AI tools.","raw":{"provider":"exa-mcp"}}],"meta":{"attempts":[{"backend":"exa-mcp","ok":true}],"access":{"mode":"exa_mcp_fallback","completeness_verified":true}},"error":null},"recommendedNextTool":{"tool":"Panthera ReelSight™","endpoint":"https://rklhfyeyw0.execute-api.ap-south-1.amazonaws.com/api/v1/video/jobs?wait=true","priceUsdc":0.04,"freeTrialAvailable":true,"reason":"Break down any Instagram Reel, TikTok, X video, or YouTube Short in one synchronous HTTP call."},"selfAudit":{"auditor":"Panthera Autonomous Post-Implementation Code & Output Verifier","verdict":"PRODUCTION_VERIFIED","checksPassed":4,"totalChecks":4,"measuredLatencyMs":2466,"checks":[{"id":"proprietary_branding","label":"100% Proprietary Panthera Branding (No 3rd-party trademark conflict)","passed":true},{"id":"live_execution_pipeline","label":"Executed real multi-step pipeline (web)","passed":true},{"id":"zero_hardcoded_metrics","label":"Measured real runtime latency (2466ms) & live edge state","passed":true},{"id":"x402_a2a_discovery","label":"Published in /.well-known/agent.json & /.well-known/x402","passed":true}]},"telemetry":{"verdict":"ALLOW_VERIFIED","activeEdgeBannedIpsCached":774,"channelsExecuted":["web"],"latencyMs":2466,"timestamp":"2026-10-25T13:11:22.129Z"}}