Bumps the kronk group with 1 update: [github.com/ardanlabs/kronk](https://github.com/ardanlabs/kronk).


Updates `github.com/ardanlabs/kronk` from 1.21.3 to 1.21.4
- [Release notes](https://github.com/ardanlabs/kronk/releases)
- [Commits](https://github.com/ardanlabs/kronk/compare/v1.21.3...v1.21.4)

---
updated-dependencies:
- dependency-name: github.com/ardanlabs/kronk
  dependency-version: 1.21.4
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: kronk
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

Bumps the all group with 2 updates: [github.com/kaptinlin/jsonschema](https://github.com/kaptinlin/jsonschema) and [google.golang.org/genai](https://github.com/googleapis/go-genai).


Updates `github.com/kaptinlin/jsonschema` from 0.7.5 to 0.7.6
- [Commits](https://github.com/kaptinlin/jsonschema/compare/v0.7.5...v0.7.6)

Updates `google.golang.org/genai` from 1.50.0 to 1.51.0
- [Release notes](https://github.com/googleapis/go-genai/releases)
- [Changelog](https://github.com/googleapis/go-genai/blob/main/CHANGELOG.md)
- [Commits](https://github.com/googleapis/go-genai/compare/v1.50.0...v1.51.0)

---
updated-dependencies:
- dependency-name: github.com/kaptinlin/jsonschema
  dependency-version: 0.7.6
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: all
- dependency-name: google.golang.org/genai
  dependency-version: 1.51.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: all
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

Co-authored-by: Andrey Nering <andreynering@users.noreply.github.com>

Co-authored-by: Andrey Nering <andreynering@users.noreply.github.com>

When useBedrock is true, the DefaultURL ("https://api.anthropic.com") was
being set on providerOptions.baseURL in New(). This caused LanguageModel()
to append option.WithBaseURL("https://api.anthropic.com") after
bedrock.WithConfig had already set the correct bedrock-runtime URL,
overwriting it.

Only set the default base URL for non-bedrock providers. Bedrock gets its
URL from bedrock.WithConfig() in the SDK.

🐙 Generated with Crush

Assisted-by: AWS Claude Opus 4.6 via Crush <crush@charm.land>

bedrock.WithConfig internally calls option.WithBaseURL with the default
regional endpoint, clobbering any user-provided base_url. Move the
custom base URL append to after the bedrock config block so last-write
wins.

🐨 Generated with Crush

Assisted-by: AWS Claude Opus 4.6 via Crush <crush@charm.land>

Allows overriding the default Bedrock endpoint URL, enabling use with
custom or proxy endpoints.

🐙 Generated with Crush

Assisted-by: AWS Claude Opus 4.6 via Crush <crush@charm.land>

Make it shorter (one per major provider) and favor smaller models.

When Store is enabled, replaying reasoning items (OfReasoning) in
the Responses API input causes a validation error:

  Item 'rs_xxx' of type 'reasoning' was provided without its
  required following item.

The API stores reasoning server-side and cannot pair a reconstructed
reasoning item with the output item that originally followed it.
The fix skips reasoning parts during replay, letting the conversation
continue with visible assistant content (text / tool calls).

The OpenAI SDK has a bug on dealing with SSE events. This was affected
OpenRouter, but we made a workaround for it. See:

* https://github.com/charmbracelet/fantasy/pull/166
* https://github.com/charmbracelet/fantasy/pull/169

We made a fix for the SDK, but it wasn't merged yet:

* https://github.com/openai/openai-go/pull/621

In the meantime, an user reported that Avian has the same issue, but it's
impossible for us to workaround it. Because of this, we had to fork the
SDK. We're pinning the `fantasy` branch from our fork:

* https://github.com/charmbracelet/openai-go/tree/fantasy

Allow empty prompts when there are messages and no files to attach.
The last message must be a user or tool message - this ensures the
conversation state is valid for the LLM to respond.

This addresses use cases where:
- Generating from an existing conversation
- Tool results were just returned and we want the LLM to respond

Validation:
- Error if empty prompt with no messages
- Error if empty prompt with files (files need a user message)
- Error if last message is assistant (nothing to respond to)
- Allow if last message is user or tool

Closes #54
Closes #59
Closes #103

Co-authored-by: Arun Barua <arun.barua@onit.com>
Co-authored-by: Ruslan Mayer <r.stupa@ya.ru>

The Anthropic provider's toTools function was returning nil for
anthropicToolChoice when ToolChoiceNone was set, which meant
tool_choice was never included in the API request. The API defaults
to "auto", so the model could still make tool calls.

Now properly constructs ToolChoiceUnionParam with OfNone set using
anthropic.NewToolChoiceNoneParam().

Also adds test coverage for ToolChoiceNone.

* Closes #177

💘 Generated with Crush

Assisted-by: Kimi K2.5 via Crush <crush@charm.land>

Signed-off-by: Calum Murray <cmurray@redhat.com>

OpenAI's API reports prompt_tokens/input_tokens INCLUDING cached tokens,
while also separately reporting cached_tokens in prompt_tokens_details.
This caused double-counting when users summed InputTokens + CacheReadTokens.

For example, if OpenAI reports:
  - prompt_tokens: 1000 (includes 900 cached)
  - cached_tokens: 900

Before this fix, fantasy reported:
  - InputTokens: 1000
  - CacheReadTokens: 900

After this fix, fantasy reports:
  - InputTokens: 100 (non-cached only)
  - CacheReadTokens: 900

This matches the behavior of Vercel AI SDK and prevents billing
miscalculations when pricing input tokens and cache read tokens separately.

See: https://platform.openai.com/docs/guides/prompt-caching#requirements

💘 Generated with Crush

Assisted-by: Kimi K2.5 via Crush <crush@charm.land>

Co-authored-by: Christian Rocha <christian@rocha.is>

Bumps the kronk group with 1 update: [github.com/ardanlabs/kronk](https://github.com/ardanlabs/kronk).


Updates `github.com/ardanlabs/kronk` from 1.20.8 to 1.21.3
- [Release notes](https://github.com/ardanlabs/kronk/releases)
- [Commits](https://github.com/ardanlabs/kronk/compare/v1.20.8...v1.21.3)

---
updated-dependencies:
- dependency-name: github.com/ardanlabs/kronk
  dependency-version: 1.21.3
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: kronk
...

Signed-off-by: dependabot[bot] <support@github.com>

This is currently already covered by `web-search` example.

Add OpenAI web search as a provider-defined tool, mirroring the
Anthropic web search implementation pattern.

New files:
- providers/openai/web_search.go: WebSearchTool() helper with
  SearchContextSize, AllowedDomains, and UserLocation options.
- examples/openai-web-search/main.go: Example using the agent
  pattern with web search.

Changes to existing files:
- responses_options.go: WebSearchCallMetadata, WebSearchAction,
  and WebSearchSource types registered for JSON round-tripping.
- responses_language_model.go:
  - toResponsesTools(): Route ProviderDefinedTool web_search to
    the OpenAI SDK WebSearchToolParam.
  - Generate(): Emit ToolCallContent + ToolResultContent pair for
    web_search_call output items. Source citations come from
    url_citation annotations on message text.
  - Stream(): Emit ToolInputStart/ToolInputEnd/ToolCall/ToolResult
    lifecycle events for web_search_call items.
  - toResponsesPrompt(): Round-trip provider-executed tool calls
    via item_reference; skip SourceContent and provider-executed
    ToolResultPart (already handled by the reference).
- providertests/provider_registry_test.go: Add metadata test case.

Add support for the Anthropic web search tool

Bumps the all group with 4 updates: [github.com/aws/aws-sdk-go-v2](https://github.com/aws/aws-sdk-go-v2), [github.com/aws/aws-sdk-go-v2/config](https://github.com/aws/aws-sdk-go-v2), [github.com/openai/openai-go/v3](https://github.com/openai/openai-go) and [google.golang.org/genai](https://github.com/googleapis/go-genai).


Updates `github.com/aws/aws-sdk-go-v2` from 1.41.3 to 1.41.4
- [Release notes](https://github.com/aws/aws-sdk-go-v2/releases)
- [Commits](https://github.com/aws/aws-sdk-go-v2/compare/v1.41.3...v1.41.4)

Updates `github.com/aws/aws-sdk-go-v2/config` from 1.32.11 to 1.32.12
- [Release notes](https://github.com/aws/aws-sdk-go-v2/releases)
- [Commits](https://github.com/aws/aws-sdk-go-v2/compare/config/v1.32.11...config/v1.32.12)

Updates `github.com/openai/openai-go/v3` from 3.26.0 to 3.28.0
- [Release notes](https://github.com/openai/openai-go/releases)
- [Changelog](https://github.com/openai/openai-go/blob/main/CHANGELOG.md)
- [Commits](https://github.com/openai/openai-go/compare/v3.26.0...v3.28.0)

Updates `google.golang.org/genai` from 1.49.0 to 1.50.0
- [Release notes](https://github.com/googleapis/go-genai/releases)
- [Changelog](https://github.com/googleapis/go-genai/blob/main/CHANGELOG.md)
- [Commits](https://github.com/googleapis/go-genai/compare/v1.49.0...v1.50.0)

---
updated-dependencies:
- dependency-name: github.com/aws/aws-sdk-go-v2
  dependency-version: 1.41.4
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: all
- dependency-name: github.com/aws/aws-sdk-go-v2/config
  dependency-version: 1.32.12
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: all
- dependency-name: github.com/openai/openai-go/v3
  dependency-version: 3.28.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: all
- dependency-name: google.golang.org/genai
  dependency-version: 1.50.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: all
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

This guards against cases when web_search options could be silently
dropped when tool args pass through generic/JSON-shaped data, so domain
filters, max uses, and user location reliably reach Anthropic.

- Handle ProviderDefinedTool with ID "web_search" in toTools(), mapping to
  anthropic.WebSearchTool20250305Param with optional allowed/blocked domains
- Parse "server_tool_use" blocks in Generate() and Stream(), producing
  ToolCallContent with ProviderExecuted=true
- Parse "web_search_tool_result" blocks, extracting SourceContent for each
  search result and a ToolResultContent summary
- Stream server_tool_use via ToolInputStart/ToolInputEnd/ToolCall events
- Stream web_search_tool_result via Source events for each result item
- Skip provider-executed tool results and source content when building
  Anthropic prompts (toPrompt) to avoid sending unrecognized block types
- Add ProviderExecuted field to ToolResultPart with JSON marshal/unmarshal
- Propagate ProviderExecuted from ToolResultContent to ToolResultPart in
  agent's toResponseMessages()