Agent Permissions
Design least-privilege access and approval boundaries for autonomous agents.
Topic guide
Go deep on agent permissions.
Read the full IndieFounder collection, from fundamentals to production implementation, security boundaries, failure modes, and operational practices.
All Agent Permissions articles
Newest and most relevant articles first. The collection grows automatically as new articles are published.
AI Agent RBAC vs ABAC: Which Authorization Model Should You Use?
RBAC is simple to operate, while ABAC can express the resource, tenant, and context rules that agent workflows often need.
AI Agent Permission Escalation: How to Stop Agents From Granting Themselves Access
Agents should never be able to turn a normal task into an authorization change without crossing a trusted control boundary.
AI Coding Agent Command Permissions: How to Control Shell Access
A coding agent with unrestricted shell access can do far more than edit source files.
MCP Tool Permissions: How to Limit What an Agent Can Access
Connecting an MCP server can expose many capabilities at once.
Multi-Tenant AI Agents: How to Prevent Cross-Tenant Access
An agent serving multiple customers must never rely on the model to choose the correct tenant. Tenant boundaries belong in trusted authorization code.
AI Agent Resource Scoping: How to Limit Access to Specific Records
Giving an agent access to a database does not mean giving it access to every row. Resource-level authorization keeps automation inside its intended boundary.
AI Agent Approval Policies: Designing Human Checkpoints by Risk
Not every tool call needs a human. Approval should be tied to the potential impact, reversibility, and uncertainty of the action.
AI Agent Permission Testing: How to Test Authorization Before Production
Permission bugs can remain invisible until an agent reaches the wrong resource. Build adversarial authorization tests before deployment.
AI Agent Just-in-Time Permissions: How to Grant Access Only When Needed
Long-lived agent permissions increase blast radius. Just-in-time access can grant a narrow capability for one workflow and remove it afterward.
AI Agent Permissions: Designing Least-Privilege Access
Design agent access like a capability system: narrow tools, scoped credentials, tenant boundaries, and separate read/write permissions.
AI Agent Permission Boundaries: How to Separate Read, Write, and Delete
A safe agent rarely needs one broad permission. Split capabilities by impact so low-risk reads do not inherit destructive access.
AI Agent Delegated Identity: How to Act on Behalf of a User Safely
An agent may perform an action for a human, but the audit trail should preserve both identities.
AI Agent Identity: Why Every Agent Needs a Distinct Security Principal
An agent should not operate as an anonymous extension of the user or as a shared service account. Give automated actors explicit identity and scoped authority.
AI Agent Architecture: Models, Tools, Memory, Permissions and Logs
A production agent is a software system around a model. Separate reasoning, tools, state, permissions, approvals, and observability.
AI Coding Agent Deployment Permissions: How to Control Production Releases
Allowing an agent to deploy code turns a coding assistant into a production operator.
AI Coding Agent Database Access: How to Protect Production Data
A coding agent that can inspect or modify a database can create a much larger blast radius than source-code access alone.
AI Agent API Authentication: OAuth vs API Keys and Service Identity
Choosing authentication for an agent is different from choosing authentication for a normal backend integration.
How to Give AI Agents Secure Access to External Tools
A practical architecture for giving AI agents useful tool access without turning the model into a production superuser.
Flatkey and the One-API Model: AI Access Is Becoming a Commodity Layer
As startups bundle access to many official AI models behind one billing and routing layer, the competitive advantage is moving upward into workflow and product experience.
MCP Multi-Tenant Security: How to Isolate Customer Data
A shared MCP server must never let one tenant's agent access another tenant's resources.
If an AI Agent Can Touch Production, Your SaaS Needs a Permission Map
You do not need a security team to keep a one-person product honest. You need a short list of controls you can still explain at 2 a.m.
How to Secure AI Agents With Database and API Access
Keep agents away from raw database superusers. Put business APIs between the model and data, then enforce tenant, role, row, and operation-level checks.
Human Approval in AI Agents: Where Should the Checkpoint Go?
Approval works best at the boundary just before a consequential side effect, with an exact operation, clear preview, and server-side verification.
OAuth vs API Keys for AI Agents: Which Should Developers Use?
A practical comparison of API keys, OAuth, and service identities for AI agents.
AI Agent Security Checklist Before Production
Use a practical pre-launch checklist covering identity, permissions, tools, secrets, sandboxing, prompt injection, approvals, logging, and rollback.
AI Agents Are Creating a New Runtime Security Layer
As software agents gain permission to use tools, access data and take actions, security is moving from model prompts into the runtime itself.
The New Developer Tooling Layer: Security for AI Agents
As coding agents gain access to repositories and external tools, security around permissions, provenance, and safe execution is becoming a new product category.
AI Agent Step-Up Authentication: When Actions Need Extra Verification
Not every agent action deserves the same authentication strength.
More from the hub
Explore another topic
AI Agent Security
Protect agents, tools, credentials, data, and execution environments.
ExploreTool Calling
Understand how agents safely call APIs, databases, browsers, and external tools.
ExploreAgent APIs
Build reliable APIs and interfaces for agents that need to take actions.
ExploreMCP
Connect models to tools and data without turning every integration into a security hole.
Explore