Content Agent: AI That Runs Content Operations at Scale
Last updated: 17.01.2026
Content Agent is an AI built specifically for large-scale content operations, going far beyond traditional AI writing assistants that only handle one document at a time. It understands your content schema—including document types, fields, validation rules, and relationships—so it can transform raw source material into structured content, audit entire libraries, and execute coordinated updates across thousands of documents. Key capabilities include pipeline transformations from sources like press releases or specs into blog posts or product pages, large-scale analysis to detect missing metadata or outdated content, intelligent bulk editing for rebrands and URL changes, visual editing of images via natural language, and integrated web research to keep content accurate and current. Technically, it leverages document sets, staged changes with reviewable releases, schema-aware validation, a specialized multi-agent architecture, and direct integration with Sanity’s Content Lake. Content Agent is included in all Sanity plans and uses an AI credit system based on queries and actions, with clear examples of typical usage and tools for monitoring and controlling costs.
Content operations at scale is hard. Managing thousands of documents, maintaining consistency across a content library, transforming source material into polished articles, and keeping everything up to date—these tasks consume enormous amounts of time and resources. Traditional AI writing assistants can help draft individual pieces, but they weren't built to understand your content structure, audit your entire library, or make coordinated changes across hundreds of documents at once.
That's where Content Agent comes in. It's the first AI agent built specifically for content operations at scale, designed to handle the complex, repetitive, and time-consuming work that content teams face every day.
Beyond AI Writing Assistants
Most AI writing tools are essentially smart text generators. They can help you write a blog post or polish a paragraph, but they operate in isolation—one document at a time, with no understanding of your content structure or how documents relate to each other.
Content Agent is fundamentally different. It understands your content schema—the document types, field requirements, validation rules, and relationships that define your content model. It can analyze your entire content library, identify patterns and problems, and execute coordinated changes across thousands of documents. It's not just writing content; it's running content operations.
Key Capabilities
1. Pipeline Capabilities: From Source to Published
Content Agent excels at transforming raw source material into polished, published content. Need to turn a press release into a blog post? Just ask: "Turn this press release into a blog post." The agent understands the structure of your blog post document type, extracts the key information from the source material, and creates a properly formatted article with all required fields populated.
This pipeline capability extends beyond simple conversion. Content Agent can take meeting notes, product specifications, research reports, or any other source material and transform it into the appropriate content format for your needs—whether that's a blog post, product page, case study, or documentation article.
2. Analysis at Scale: Content Health and Gap Detection
One of Content Agent's most powerful capabilities is its ability to audit and analyze content at scale. You can ask questions like "Find all posts missing meta descriptions and write them" or "Which articles need updating based on our new product launch?" The agent can scan through thousands of documents, identify issues, gaps, and opportunities, and even fix them automatically.
This capability is invaluable for maintaining content health. Instead of manually reviewing documents one by one, you can get instant insights into your entire content library: missing SEO metadata, outdated information, broken references, inconsistent formatting, or content gaps in your coverage.
3. Bulk Editing: Coordinated Changes Across Hundreds of Documents
When your company rebrands, updates a product name, or needs to change URLs across your entire site, Content Agent can handle it. It can update hundreds or thousands of documents at once, making coordinated changes while respecting your content structure and validation rules.
The bulk editing capability goes beyond simple find-and-replace. Content Agent understands context and can make intelligent updates. For example, it can update all references to an old product name while preserving the surrounding text, adjust URLs while maintaining proper slug formatting, or update brand terminology consistently across different content types.
4. Visual Editing: Adjusting Images with Natural Language
Content isn't just text. Content Agent can generate and edit images using natural language instructions. Need to adjust the hero image on a landing page? Just describe what you want changed. The agent can generate new images from scratch or modify existing ones, all while properly managing image assets and metadata within your content structure.
5. Research Integration: Connecting Internal Content with External Intelligence
Content Agent doesn't operate in a vacuum. It can perform web research to gather current information, market trends, and competitive intelligence, then integrate that research with your internal content. This capability is essential for creating content that's not only well-structured but also accurate, current, and informed by the latest developments in your industry.
Technical Features That Make It Powerful
Several technical capabilities distinguish Content Agent from typical AI tools and make it uniquely suited for content operations at scale.
Document Sets and Bulk Operations
Content Agent can work with document sets—collections of related documents that need coordinated updates. This enables true bulk operations where changes are applied consistently across multiple documents while maintaining referential integrity and respecting validation rules.
Staged Changes and Releases
One of the most important features for enterprise content operations is the ability to review changes before they go live. Content Agent supports staged changes—you can review all modifications, approve or reject them, and then commit them as a coordinated release. This gives you control and confidence when making large-scale updates.
Schema-Aware Operations
Content Agent understands your content schema deeply. It knows which fields are required, what validation rules apply, which document types can reference each other, and how your content model is structured. This schema awareness ensures that all changes respect your content model and maintain data integrity.
When creating or updating documents, Content Agent automatically validates field types, respects maximum lengths, ensures required fields are populated, and maintains proper reference relationships. It's not just making changes—it's making valid, structurally correct changes.
Multi-Agent Architecture
Under the hood, Content Agent uses a multi-agent architecture with specialized agents for different tasks. There are agents optimized for document mutations, query operations, image generation, and research. This specialization enables more efficient and accurate operations—each agent is purpose-built for its specific domain.
Direct Content Lake Integration
Content Agent operates directly on Sanity's Content Lake, the underlying data store for all your content. This direct integration means the agent has full access to your content structure, can perform complex queries efficiently, and can make updates that are immediately reflected across all your applications and channels.
Pricing and Availability
Content Agent is included in all Sanity plans—Free, Growth, and Enterprise. Rather than charging per seat or per feature, Content Agent uses an AI credit system that aligns costs with actual usage.
How AI Credits Work
Content Agent operations consume AI credits based on two types of activities:
- Query (your message to Content Agent): 4 credits
- Action (tool use by Content Agent): 2 credits
Actions include document analysis, content creation, web searches, and GROQ queries. For multi-document operations, credits are counted per document—so analyzing 10 articles means 10 separate actions.
Monthly Included Credits
Each Sanity plan includes a monthly allocation of AI credits:
- Free plan: 100 AI credits per month
- Growth plan: 100 AI credits per month
- Enterprise plan: 500 AI credits per month
If you need more credits, additional capacity is available. Growth plan customers can purchase credits at $0.05 per AI credit, while Enterprise customers have access to AI credit bundles tailored to their usage patterns.
Understanding Credit Consumption
To help you understand how credits translate to real work, here are some practical examples:
- "Show Q3 blog posts" = 1 query + 1 action = 6 credits
- "Analyze 10 articles for SEO issues" = 1 query + ~12 actions = ~28 credits
- "Create 5 product pages from specifications" = 1 query + ~15 actions = ~34 credits
The exact number of actions varies based on the complexity of the task and the number of documents involved, but this gives you a sense of how credits map to actual content operations.
Cost Control and Monitoring
Sanity provides comprehensive tools to help you manage AI credit usage. You can set spending limits to prevent unexpected charges, monitor usage in real-time through your dashboard, and pause AI features if needed. This gives you full control over costs while still benefiting from Content Agent's powerful capabilities.
AI That Actually Runs Content Operations
Content Agent represents a fundamental shift in how AI can support content teams. This isn't AI bolted onto a CMS as an afterthought—it's AI that actually runs content operations. It understands your content model, works at scale, maintains data integrity, and gives you control through staged releases and review workflows.
For content teams managing large libraries, Content Agent transforms what's possible. Tasks that would take days or weeks—auditing thousands of documents, making coordinated updates, transforming source material into published content—can now be accomplished in minutes. And because it's schema-aware and supports staged changes, you maintain full control and confidence in the results.
This is content operations at scale, powered by AI that truly understands content.
When the Claude window runs out: my take on T3 Code
T3 Code addresses the frustration of hitting Claude’s five-hour usage limit during focused work. Rather than replacing coding agents, it provides a single desktop interface for tools such as Claude Code, Codex, Cursor, Grok Build, OpenCode, and Antigravity, with users signing in through their own accounts. Each thread remains tied to its original provider, so an unfinished Claude conversation cannot be transferred directly to Codex. However, the surrounding workspace, project, and workflow stay in one app. When a limit is reached, T3 Code identifies the restriction and reset time, allowing the user to start a new thread with another provider and quickly restate the task. The app is free and does not resell or mark up tokens, aligning its incentives with users who want to avoid unnecessary calls. Its MIT license, self-hosting support, public codebase, and more than 21,000 GitHub stars are especially important because it handles provider credentials and repository access. Although founder Theo frequently promotes the product, its transparency makes those claims easier to verify. The Linux desktop app remains young and occasional releases cause problems, but steady improvements and its ability to coordinate separate agents make it a practical solution for maintaining momentum.
Read post
Trying out autonomous agents
Sandcastle is Matt Pocock’s workflow for running Claude Code unattended inside a Docker container. It assigns the agent a GitHub issue in a git worktree, then merges the resulting commits back into the user’s branch. Its init command quickly scaffolds the infrastructure, but useful results still depend on supplying clear, focused tickets. Wayfinder fills that planning gap by breaking an idea that is too large for one session into a map issue and smaller decision tickets. Each ticket addresses a single unresolved question rather than a build segment, producing the self-contained, labeled issues Sandcastle needs. Without this map, vague issues led to equally vague pull requests. Before queueing work, the author uses a grilling skill to challenge assumptions and force difficult decisions. Resolving uncertainty early prevents the agent from making costly guesses during unattended runs. For secrets, the author replaces a plaintext .env file with Varlock and a committed .env.schema. Values are resolved through Doppler and exposed only through process.env during execution, although each secret must still be explicitly forwarded into the container configuration. The author would use the setup again but stresses that ticket quality remains decisive. Every diff is still reviewed, tasks stay tightly scoped, and fully unattended overnight operation remains a future goal.
Read post