AI agents

Scope manifests and tool permissions

• A Tool Registry With Tiers: Who May Call What
• Per-Caller Caps Layered on Top of Tool Tiers
• Per-Caller Caps Over Tool Tiers: Weights and a Floor
• Permission Tests That Catch a Widened Subagent Tool
• Subagent Tool Scoping: Limiting What a Delegated Task Can Touch
• Tier Claims in the Token Versus a Lookup per Call
• A Clearance Check at Execution: Wrapping Every Tool Call
• A Scope Manifest per Agent: Verbs, Resources and Owners
• A Token That Carries Tier: Fields, TTL and a Signature Check
• An Audit Query: Who Called This Tool, Under Which Tier
• Blast Radius Calculation for Agent Actions
• Credential Scope per Tool: What the Agent Should Never Hold
• Describing a Tool When the Tier Is Also a Rule
• Fixtures That Catch a Widened Scope in a Delegation Chain
• Least Privilege for Agents: Scoping Tools So Injections Do Little
• Narrowing a Tool's Tier Without an Unguarded Gap
• Read Permissions per Retriever: Who May See a Held Chunk
• Revocation Lag After a Downgrade: Budgeting the Gap
• Tier Checks in the Tool Schema: A Second Line of Defense
• Tool Permissions by Role: Limiting What an Agent Can Call
• What a Refusal Log Must Cite to Survive an Audit
• When a Tool Tier Changes: Migrating Callers Without a Gap