--- title: Agent Run Observability in Laravel AI SDK 0.11 | daily.dev description: Laravel AI SDK 0.11.0 introduces run-level observability for agent workflows: a single correlation ID threaded through an entire run, and lifecycle events... canonical: https://daily.dev/posts/agent-run-observability-in-laravel-ai-sdk-0-11-2ws8tt7cm twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Agent Run Observability in Laravel AI SDK 0.11 | daily.dev og:description: Laravel AI SDK 0.11.0 introduces run-level observability for agent workflows: a single correlation ID threaded through an entire run, and lifecycle events... og:url: https://daily.dev/posts/agent-run-observability-in-laravel-ai-sdk-0-11-2ws8tt7cm og:image: https://api.daily.dev/og/posts/2wS8tT7Cm.png og:image:alt: Agent Run Observability in Laravel AI SDK 0.11 og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Agent Run Observability in Laravel AI SDK 0.11 **[Laravel News](https://daily.dev/sources/ln)** · 8 min read · 0 upvotes · 0 comments ## Summary Laravel AI SDK 0.11.0 introduces run-level observability for agent workflows: a single correlation ID threaded through an entire run, and lifecycle events (StartingStep, StepCompleted, StepFailed, ToolFailed, AgentFailed) with wall-clock timings. Hosted tool search now works for OpenAI and Anthropic via a ToolSearch wrapper, sparing token cost from shipping full tool catalogues on every request. Failover expands to cover connection failures, more upstream HTTP error codes (502/504/520/522/524), and Anthropic usage-cap errors. Other additions include xAI web/file search, Groq and OpenAI-compatible transcription, an assertPromptedTimes() test helper, and Gemini's default text model moving to gemini-3.7-flash. The release shipped August 19, 2026 with 36 merged PRs and twelve first-time contributors, and includes upgrade notes for two events with new required constructor arguments. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What does Laravel AI SDK 0.11 add for tracing agent runs? Laravel AI SDK 0.11.0 adds a single correlation ID threaded through an entire agent run, including failover attempts, along with lifecycle events StartingStep, StepCompleted, StepFailed, ToolFailed, and AgentFailed that carry wall-clock timings around every provider round-trip and tool call. Previously, prompt() did not mint an invocation ID while streamPrompt() did, and nested tool calls could overwrite each other's IDs. _daily.dev surfaces release notes like this for developers instrumenting AI agent pipelines._ ### How does the ToolSearch wrapper in Laravel AI SDK reduce token usage for agent tool calls? The ToolSearch wrapper, added in Laravel AI SDK 0.11, defers a group of tools so OpenAI and Anthropic load them on demand via their own hosted tool search instead of shipping every tool definition on every request. It requires no interface changes to the wrapped tools, supports a regex or bm25 strategy on Anthropic, and requires OpenAI stored responses (store=false throws). _Developers weighing token costs in agent tool catalogues can track changes like this via daily.dev._ ### Which HTTP status codes trigger provider failover in Laravel AI SDK 0.11, and why was 500 excluded? Laravel AI SDK 0.11 expanded failover triggering statuses from just 503 to 502, 503, 504, 520, 522, and 524, covering gateway timeouts and Cloudflare errors from providers behind Cloudflare. A bare 500 was deliberately excluded because it can represent a deterministic error that failover would incorrectly mask rather than a transient outage. _Teams tuning failover logic for AI providers can follow releases like this on daily.dev._ ## Similar posts on daily.dev - [Laravel AI: Trace Agent Runs With Lifecycle Events](https://daily.dev/posts/laravel-ai-trace-agent-runs-with-lifecycle-events-k7bbbwg9b) · Laravel News · 2 upvotes · 0 comments - [Your Agent Harness Needs Runtime Security](https://daily.dev/posts/your-agent-harness-needs-runtime-security-bdrnogl9c) · Daily Dose of Data Science \| Avi Chawla \| Substack · 1 upvotes · 0 comments - [Laravel AI: Load Tools On Demand With ToolSearch](https://daily.dev/posts/laravel-ai-load-tools-on-demand-with-toolsearch-souygqalt) · Laravel News · 2 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#php](https://daily.dev/tags/php), [#laravel](https://daily.dev/tags/laravel), [#observability](https://daily.dev/tags/observability) [View this post on daily.dev](https://daily.dev/posts/agent-run-observability-in-laravel-ai-sdk-0-11-2ws8tt7cm) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Agent Run Observability in Laravel AI SDK 0.11","url":"https://daily.dev/posts/agent-run-observability-in-laravel-ai-sdk-0-11-2ws8tt7cm","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/agent-run-observability-in-laravel-ai-sdk-0-11-2ws8tt7cm"},"datePublished":"2026-08-21T13:47:22.562Z","dateModified":"2026-09-13T19:20:58.267Z","description":"Laravel AI SDK 0.11.0 introduces run-level observability for agent workflows: a single correlation ID threaded through an entire run, and lifecycle events...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/6ec33ffd5d5a57d7f66abc4c92da63d3?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/6ec33ffd5d5a57d7f66abc4c92da63d3?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Laravel News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Laravel News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/ln","url":"https://daily.dev/sources/ln"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/agent-run-observability-in-laravel-ai-sdk-0-11-2ws8tt7cm","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,php,laravel,observability","timeRequired":"PT8M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Laravel News","item":"https://daily.dev/sources/ln"},{"@type":"ListItem","position":3,"name":"Agent Run Observability in Laravel AI SDK 0.11"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/agent-run-observability-in-laravel-ai-sdk-0-11-2ws8tt7cm#faq","mainEntity":[{"@type":"Question","name":"What does Laravel AI SDK 0.11 add for tracing agent runs?","acceptedAnswer":{"@type":"Answer","text":"Laravel AI SDK 0.11.0 adds a single correlation ID threaded through an entire agent run, including failover attempts, along with lifecycle events StartingStep, StepCompleted, StepFailed, ToolFailed, and AgentFailed that carry wall-clock timings around every provider round-trip and tool call. Previously, prompt() did not mint an invocation ID while streamPrompt() did, and nested tool calls could overwrite each other's IDs. daily.dev surfaces release notes like this for developers instrumenting AI agent pipelines."}},{"@type":"Question","name":"How does the ToolSearch wrapper in Laravel AI SDK reduce token usage for agent tool calls?","acceptedAnswer":{"@type":"Answer","text":"The ToolSearch wrapper, added in Laravel AI SDK 0.11, defers a group of tools so OpenAI and Anthropic load them on demand via their own hosted tool search instead of shipping every tool definition on every request. It requires no interface changes to the wrapped tools, supports a regex or bm25 strategy on Anthropic, and requires OpenAI stored responses (store=false throws). Developers weighing token costs in agent tool catalogues can track changes like this via daily.dev."}},{"@type":"Question","name":"Which HTTP status codes trigger provider failover in Laravel AI SDK 0.11, and why was 500 excluded?","acceptedAnswer":{"@type":"Answer","text":"Laravel AI SDK 0.11 expanded failover triggering statuses from just 503 to 502, 503, 504, 520, 522, and 524, covering gateway timeouts and Cloudflare errors from providers behind Cloudflare. A bare 500 was deliberately excluded because it can represent a deterministic error that failover would incorrectly mask rather than a transient outage. Teams tuning failover logic for AI providers can follow releases like this on daily.dev."}}]} ```