AI Agents That Work With Grok (2026)
As of Aug 11, 2026, 11 AI agents in the KanonAgent index are judged to connect to Grok; this page covers the top 9 by real traction (9 with the source line quoted). Judgements come from quoted evidence, not vendor claims — where we cannot read it, we leave it blank.
Connecting an agent to Grok unlocks direct model access for real-time decisions, content generation, and workflow orchestration without switching platforms. Evidence shows OAuth, MCP servers, and native skill integrations are the dominant patterns in 2026. This lets users run Grok inside specialized tools rather than generic chat interfaces.
Updated 2026-08-11 · 9 products · live data from KanonAgent
1. Orus122 upvotes
Orus analyzes perpetuals markets and executes trades; the Grok OAuth link lets it pull live data and issue orders inside the same authenticated session.
source: “Connect WhatsApp, Telegram, Claude, ChatGPT, or Grok via OAuth”
2. Ycode AI Agents102 upvotes
Ycode AI Agents edit websites from natural language; Grok connection supplies an extra model option for generating or revising site content without code.
source: “Connect Claude, OpenAI, Gemini or Grok”
3. Use Any Agent as an Orchestrator10 upvotes
Use Any Agent as an Orchestrator coordinates multi-agent workflows; explicit Grok support lets it treat Grok as one of the callable executors in the pipeline.
source: “Codex · Claude Code · Copilot CLI · Grok”
4. t0md9 upvotes
t0md converts documents to Markdown for AI use; its MCP server exposes the output directly to Grok so agents can ingest structured files immediately.
source: “MCP server included for Claude Code, Cursor, Grok and other AI agents.”
5. PinCreatorTool5 upvotes
PinCreatorTool turns product photos into short vertical videos; Grok Imagine integration supplies the generation step inside the ad-creation flow.
source: “for TikTok, Reels, and Shorts with Grok Imagine”
6. Batuk — Sovereign AI for Teams4 upvotes
Batuk runs multiple models in one secure workspace; adding Grok gives teams a single authenticated endpoint for private, policy-controlled access.
source: “Connect Ollama, OpenAI, Claude, Grok, OpenRouter, Sarvam AI, or private OpenAI-compatible models”
Byre provides a low-friction agent sandbox; Grok skill support lets users test and iterate agents that call Grok without extra setup.
source: “agent skills for Claude Code, Codex, Gemini, Grok, and OpenCode”
8. Lingera2 upvotes
Lingera saves conversations with one click; Grok support pulls threads from Grok chats into the same searchable local archive.
source: “buried across ChatGPT, Claude, Grok, Gemini, Perplexity & DeepSeek”
9. 100 Questions0 upvotes
100 Questions runs brand-visibility benchmarks; it includes Grok in the 25-question suite so marketers can compare mention rates across models.
source: “runs the same 25-question benchmark across OpenAI, Claude, Gemini, and Grok”
Before you wire it up
Check whether the agent uses OAuth, MCP, or direct API keys for Grok, because each changes permission scope and revocation ease. Prefer agents that list Grok explicitly in their connection string rather than generic LLM support. Watch rate limits on Grok’s image and long-context endpoints, which commonly throttle video or document-heavy workflows first. Test whether the agent handles Grok’s tool-calling format; many break on function schemas that other models accept.
FAQ
How do I authenticate an agent to Grok without exposing my xAI key?
Use the OAuth flows shown in Orus and Batuk or the MCP server pattern in t0md and 29437 so the agent never sees the raw key.
Which agents support Grok Imagine for image or video generation?
PinCreatorTool (17024) calls Grok Imagine directly; other agents treat Grok as a text model only.
What usually breaks when wiring an agent to Grok?
Tool-calling schema mismatches and context-window limits on long Markdown or benchmark runs are the most frequent failures.
How is this list ranked?
By real traction from our index (upvotes / MRR / growth rate, whichever the product actually has) — not editorial picks, and we do not accept paid placement. Sources: ProductHunt, Hacker News, GitHub, HuggingFace, Reddit, TrustMRR.
What is the inclusion bar?
A page is published only when at least 5 real products qualify. Judgement fields (autonomy, prerequisites, cost, integrations) all require a source quote — where we cannot read it, we leave it blank rather than guess.
How often is this updated?
Collection runs continuously; this page was regenerated on 2026-08-11. Every number is verifiable through our public read-only API: https://kanonagent.com/data