<?xml version='1.0' encoding='utf-8'?>
<rss version="2.0"><channel><title>BotBento field notes</title><link>https://botbento.com/blog/</link><description>Practical, sourced explanations for working with AI agents</description><language>en</language><item><title>What Actually Happens When Your Bot Cancels an MCP Tool Call?</title><link>https://botbento.com/blog/mcp-tool-call-cancellation-what-happens/</link><guid>https://botbento.com/blog/mcp-tool-call-cancellation-what-happens/</guid><description>MCP cancellation asks a server to stop work, but it may still finish. Learn the current stdio and HTTP rules and how to check side effects.</description><pubDate>Thu, 01 Oct 2026 00:00:00 +0000</pubDate></item><item><title>How Do You Scope an API Key So a Bot Can't Exceed Its Job?</title><link>https://botbento.com/blog/bot-api-key-scoping-least-privilege/</link><guid>https://botbento.com/blog/bot-api-key-scoping-least-privilege/</guid><description>A practical Stripe and GitHub example for limiting a bot's credential to its job, with the permissions to verify and the failure paths to test.</description><pubDate>Thu, 01 Oct 2026 00:00:00 +0000</pubDate></item><item><title>MCP Roots Is Deprecated. Where Should Your Bot's File Boundaries Live Now?</title><link>https://botbento.com/blog/mcp-roots-deprecated-migration/</link><guid>https://botbento.com/blog/mcp-roots-deprecated-migration/</guid><description>MCP Roots is deprecated in the 2026-07-28 spec. Learn what remains supported, why roots are advisory, and how to migrate file scoping safely.</description><pubDate>Wed, 30 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How Do You Know Your Bot's Recurring Routine Has Gone Silent?</title><link>https://botbento.com/blog/bot-routine-silent-failure-detection/</link><guid>https://botbento.com/blog/bot-routine-silent-failure-detection/</guid><description>A scheduled bot can miss its start, fail to reach its target, or stall after starting. Each silence needs a different signal and a clear recovery path.</description><pubDate>Wed, 30 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How Long Should You Let a Bot's Tool Call Run Before Timing Out?</title><link>https://botbento.com/blog/tool-call-timeout-duration/</link><guid>https://botbento.com/blog/tool-call-timeout-duration/</guid><description>A practical guide to MCP tool-call timeouts: distinguish protocol guidance from client defaults, measure real durations, and use asynchronous jobs for slow work.</description><pubDate>Tue, 29 Sep 2026 00:00:00 +0000</pubDate></item><item><title>MCP Tools Can Declare an Output Schema. Should Your Bot Validate It?</title><link>https://botbento.com/blog/mcp-tool-output-schema-validation/</link><guid>https://botbento.com/blog/mcp-tool-output-schema-validation/</guid><description>MCP tools can declare an outputSchema and return structuredContent. Here is how to validate successful results without confusing tool errors with transport failures.</description><pubDate>Tue, 29 Sep 2026 00:00:00 +0000</pubDate></item><item><title>MCP Dropped Sessions. How Do You Keep State Across Tool Calls Now?</title><link>https://botbento.com/blog/mcp-stateless-cross-call-state/</link><guid>https://botbento.com/blog/mcp-stateless-cross-call-state/</guid><description>The MCP 2026-07-28 spec removed session IDs and the initialize handshake. Here's the explicit-handle pattern that replaces them, and what breaks if you don't update.</description><pubDate>Mon, 28 Sep 2026 00:00:00 +0000</pubDate></item><item><title>What Should a Bot Do When a Multi-Step Task Fails Halfway?</title><link>https://botbento.com/blog/bot-multi-step-task-failure-rollback/</link><guid>https://botbento.com/blog/bot-multi-step-task-failure-rollback/</guid><description>A bot that books a flight then fails to book a hotel has left the world in a half-finished state. Here is how to design the undo step.</description><pubDate>Mon, 28 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How Do You Actually Revoke a Bot's Access to a Tool It No Longer Uses?</title><link>https://botbento.com/blog/revoke-bot-tool-access/</link><guid>https://botbento.com/blog/revoke-bot-tool-access/</guid><description>OAuth token revocation can have a propagation delay. Learn what the standard guarantees, how validation affects the result, and how to verify a bot has stopped using a tool.</description><pubDate>Sun, 27 Sep 2026 00:00:00 +0000</pubDate></item><item><title>MCP Tool Results: How Do You Know a Call Actually Succeeded?</title><link>https://botbento.com/blog/mcp-tool-result-success-signal/</link><guid>https://botbento.com/blog/mcp-tool-result-success-signal/</guid><description>Read MCP protocol errors, isError, structuredContent and real-world readback separately before telling a user a tool action succeeded.</description><pubDate>Sun, 27 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How Do You Verify an AI Agent Actually Finished the Task?</title><link>https://botbento.com/blog/verify-ai-agent-task-completion/</link><guid>https://botbento.com/blog/verify-ai-agent-task-completion/</guid><description>A bot saying 'done' is not proof. Here is how to check completion independently, with a worked example and what evidence to keep.</description><pubDate>Sat, 26 Sep 2026 00:00:00 +0000</pubDate></item><item><title>When Should a Bot Use a Supervisor Agent Instead of One Agent With Tools?</title><link>https://botbento.com/blog/supervisor-agent-vs-single-agent-tools/</link><guid>https://botbento.com/blog/supervisor-agent-vs-single-agent-tools/</guid><description>A concrete rule for choosing between one agent with many tools and a supervisor that delegates to specialist agents, with a worked example.</description><pubDate>Sat, 26 Sep 2026 00:00:00 +0000</pubDate></item><item><title>MCP Sampling Is Deprecated. What Should Bot Builders Do Now?</title><link>https://botbento.com/blog/mcp-sampling-deprecated/</link><guid>https://botbento.com/blog/mcp-sampling-deprecated/</guid><description>MCP's sampling feature, which let servers ask your client to run an LLM call, is now deprecated. Here is what that means for bots built on it.</description><pubDate>Fri, 25 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How Should an AI Bot Handle a Tool's Rate Limit Error?</title><link>https://botbento.com/blog/ai-bot-rate-limit-error-handling/</link><guid>https://botbento.com/blog/ai-bot-rate-limit-error-handling/</guid><description>Learn how an AI bot should classify rate limits, honor Retry-After, retry safely with jitter and a cap, and report an unfinished tool action.</description><pubDate>Fri, 25 Sep 2026 00:00:00 +0000</pubDate></item><item><title>What Can an MCP Tool Ask You For Mid-Task, and What Can't It?</title><link>https://botbento.com/blog/mcp-elicitation-mid-task-permissions/</link><guid>https://botbento.com/blog/mcp-elicitation-mid-task-permissions/</guid><description>MCP elicitation lets a connected tool pause and ask a question mid-task. Here's what the spec allows it to request, and what it must route elsewhere.</description><pubDate>Thu, 24 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How Do You Stop a Retried Tool Call From Running Twice?</title><link>https://botbento.com/blog/agent-tool-retry-idempotency-keys/</link><guid>https://botbento.com/blog/agent-tool-retry-idempotency-keys/</guid><description>A timed-out tool call can secretly succeed. See why MCP's idempotentHint doesn't protect you, and what an idempotency key actually does.</description><pubDate>Thu, 24 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How Do You Review an AI-Generated Research Note?</title><link>https://botbento.com/blog/reviewing-ai-generated-research-notes/</link><guid>https://botbento.com/blog/reviewing-ai-generated-research-notes/</guid><description>A practical checklist for checking source ownership, dates, contradictions and unresolved questions in AI-drafted research notes before you rely on them.</description><pubDate>Wed, 23 Sep 2026 00:00:00 +0000</pubDate></item><item><title>MCP Resource Indicators: What Your Bot's OAuth Tokens Must Specify</title><link>https://botbento.com/blog/mcp-resource-indicator-requirement/</link><guid>https://botbento.com/blog/mcp-resource-indicator-requirement/</guid><description>MCP's HTTP authorization spec requires OAuth resource indicators. Learn what they protect, where implementation can fail, and what to check before connecting a bot.</description><pubDate>Wed, 23 Sep 2026 00:00:00 +0000</pubDate></item><item><title>Plugin or authorized connection: what's actually different?</title><link>https://botbento.com/blog/plugin-vs-authorized-connection/</link><guid>https://botbento.com/blog/plugin-vs-authorized-connection/</guid><description>Installing a tool on a bot and letting it act on your data are two separate steps. Here is where the line sits, with a worked example.</description><pubDate>Sun, 20 Sep 2026 00:00:00 +0000</pubDate></item><item><title>Why we are building BotBento around people, not just bots</title><link>https://botbento.com/blog/people-and-bots-in-one-room/</link><guid>https://botbento.com/blog/people-and-bots-in-one-room/</guid><description>Grok Bot's group chats hold bots only. BotBento is built so people and bots share one room under one permission model. Here is the reasoning and the honest status.</description><pubDate>Sun, 20 Sep 2026 00:00:00 +0000</pubDate></item><item><title>How do you test an AI agent that uses a calendar?</title><link>https://botbento.com/blog/test-calendar-ai-agent/</link><guid>https://botbento.com/blog/test-calendar-ai-agent/</guid><description>Test a calendar AI agent with a disposable event, explicit time zone, denied write and revoked connection. Check the saved result before trusting its reply.</description><pubDate>Thu, 10 Sep 2026 00:00:00 +0000</pubDate></item><item><title>What belongs in an AI agent stopping rule?</title><link>https://botbento.com/blog/ai-agent-stopping-rules/</link><guid>https://botbento.com/blog/ai-agent-stopping-rules/</guid><description>Define when an AI agent should finish, pause or stop at a limit, with an illustrative research task and practical checks for loops and unresolved work.</description><pubDate>Thu, 10 Sep 2026 00:00:00 +0000</pubDate></item><item><title>Local or cloud: where should your AI agent run?</title><link>https://botbento.com/blog/local-vs-cloud-ai-agents/</link><guid>https://botbento.com/blog/local-vs-cloud-ai-agents/</guid><description>Choose where an AI agent runs by separating its model, tools and scheduler. Compare local, cloud and hybrid setups with a practical decision checklist.</description><pubDate>Tue, 08 Sep 2026 00:00:00 +0000</pubDate></item><item><title>What should an AI agent record after each run?</title><link>https://botbento.com/blog/ai-agent-run-result-record/</link><guid>https://botbento.com/blog/ai-agent-run-result-record/</guid><description>Design a useful AI agent run record with evidence, partial results, retry decisions and clear next steps, using an illustrative research routine.</description><pubDate>Tue, 08 Sep 2026 00:00:00 +0000</pubDate></item><item><title>MCP tool permissions: what to check before connecting a bot</title><link>https://botbento.com/blog/mcp-tool-permissions/</link><guid>https://botbento.com/blog/mcp-tool-permissions/</guid><description>Understand MCP servers, authorization and tool approval with an illustrative calendar example and a practical connection review checklist.</description><pubDate>Mon, 07 Sep 2026 00:00:00 +0000</pubDate></item><item><title>AI agents vs workflows: choose who decides the next step</title><link>https://botbento.com/blog/ai-agents-vs-workflows/</link><guid>https://botbento.com/blog/ai-agents-vs-workflows/</guid><description>A practical way to choose between an AI agent and a fixed workflow, with a weekly research example, stopping rules and an evaluation checklist.</description><pubDate>Mon, 07 Sep 2026 00:00:00 +0000</pubDate></item></channel></rss>
