* docs: add agent readability improvements
- Add AGENTS.md skill file served at /AGENTS.md
- Add /.well-known/agent-skills/ discovery index and SKILL.md
- Add JSON-LD structured data to homepage
- Exclude AGENTS.md and .well-known from i18n middleware
* fix: suppress biome lint for JSON-LD dangerouslySetInnerHTML
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
- Added a `details` field to the `task_update` type for providing additional context in task updates.
- Updated relevant documentation and test cases to reflect the new field, improving clarity on task progress reporting.
* feat(chat): add chat.thread() method for creating thread handles
Allows constructing a Thread handle from a thread ID outside of webhook
contexts, enabling proactive messaging to existing threads.
Closes#148
* docs: add chat.thread() to API reference
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* Support passing scopes as a config to the linear adapter
* Changeset
* Add multi-tenant support for Linear
* Lint
* Cleanup
* Fix tests
* Make organizationId mandatory
* Fix
* Fix fetch of comments
* Start agent sessions support
* Continue cleanup
* Cleanup more
* Simplify more
* Fix
* Fix fetchMessages for agent sessions
* Improve streaming to linear
* Fix agent raw messages
* Fix tests
* Lint
* Simplify
* Work
* Remove debug logs and adapt tests
* Fix organizationId
* Cleanup types
* Cleanup and comment
* Fix webhook handling
* Expose prompt context
* First review pass
* Rework the changeset
* Improve getInstallation and expose refreshInstallation
* Fix error on agent sessions coming from issue delegation
---------
Co-authored-by: <dancer@users.noreply.github.com>
* fix(gchat): clear cards when editing a message to text
* add changeset for gchat clear cards fix
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* feat: add apiUrl config option to all platform adapters
Add a consistent `apiUrl` option across all 8 adapters for custom API
endpoint configuration (e.g. GovSlack, GitHub Enterprise, GCC-High
Teams). Includes env var fallbacks and unit tests for each adapter.
* add changeset for adapter apiUrl feature
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* [chat] enable updating task blocks by ID, rather than only most recent
* add changeset for updateTask by ID
* test(chat): add tests for updateTask by ID
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
Co-authored-by: dancer <josh@afterima.ge>
* docs: misc updates and fixes
* docs: update feature matrix to show Teams modal support
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* fix(chat): export standalone reviver for workflow-safe deserialization
Importing the Chat instance into Vercel Workflow files pulls in adapter
packages that depend on Node.js modules, which aren't available in the
workflow sandbox. This adds a standalone `reviver` function exported from
`chat` that deserializes Thread/Channel/Message objects without needing
a Chat instance or its adapter dependencies.
Also updates the durable chat sessions guide to use the standalone
reviver and dynamic bot imports inside step functions.
Closes#243
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(chat): preserve adapterName on deserialized toJSON and add standalone reviver docs
* fix: sort imports in index.ts
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: dancer <josh@afterima.ge>
Co-authored-by: Ben Sabic <27636870+bensabic@users.noreply.github.com>
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* feat(chat): Enable disabling code-fencing for streamed tables per-adapter
As Slack now natively supports markdown tables, the code-fencing for
tables is not required and in fact stops them from being rendered
natively.
This PR adds an option to configure code-fencing tables per adapter.
By default true, false only for Slack.
* Update StreamingMarkdownRendererOptions comments
Removed comment about default behavior for wrapTablesForAppend option.
* feat(adapter-teams): add Select and RadioSelect support in card Actions
Map Select and RadioSelect card elements to Teams Adaptive Card
ChoiceSetInput (compact and expanded styles). Auto-inject a submit
button when inputs exist without explicit buttons, and fan out the
auto-submit payload into individual onAction calls per input.
* [chat] adding plan/update plan blocks for threads
* refactor plan impl into its own file
* thread.post() instead of thread.postPlan()
* fixes
* lint fixes
* refactor postable objects
* knip
* Refactor PostableObject into dedicated module and add fallback support (#191)
* feat: refactor adapter-slack and types based on Malte's comments
Move PostableObject to dedicated file, update type exports, add fallback posts, and make PostableObject generic.
Slack-Thread: https://vercel.slack.com/archives/C0977L169MW/p1772755978706959?thread_ts=1772755978.706959&cid=C0977L169MW
Co-authored-by: Vishal Yathish <135551666+visyat@users.noreply.github.com>
* fix: fallback path for PostableObject and clean up exports
- Plan mutations (addTask, updateTask, complete) now work in fallback
mode by using editMessage with fallback text instead of no-oping
- Revert Slack adapter to throw on unsupported kinds (caller handles
fallback, so adapter-level fallback strings are unnecessary)
- Remove redundant re-exports of isPostableObject/PostableObject from
plan.ts — export from canonical postable-object.ts only
- Stop exporting POSTABLE_OBJECT symbol (internal implementation detail)
- Fix nested ternary lint error in getFallbackText
- Add tests for fallback post/edit path
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: v0 <v0[bot]@users.noreply.github.com>
Co-authored-by: Hayden Bleasel <hello@haydenbleasel.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* Fall back to plain text instead of throwing for unsupported postable object kinds
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(chat): deduplicate PostableObject handling, improve type safety and test coverage
---------
Co-authored-by: v0 <v0[bot]@users.noreply.github.com>
Co-authored-by: Hayden Bleasel <hello@haydenbleasel.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: dancer <josh@afterima.ge>
* feat(github): add support for issue comment threads
Handle issue_comment webhooks on plain issues, not just PRs.
Issue threads use the format github:owner/repo:issue:42.
* fix(github): propagate threadType in GitHubRawMessage for issue threads
parseMessage and other methods now correctly produce issue-format thread
IDs when the raw message originated from a plain issue, not a PR.
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* feat(teams): add dialog (task module) support
Teams dialogs require modal content to be returned inline in the HTTP
response when a task/fetch invoke fires. This adds:
- `actionType: "modal"` on buttons to emit msteams task/fetch hint
- `onOpenModal` hook on WebhookOptions for inline modal interception
- dialog.open/dialog.submit handlers in Teams adapter with Promise.race
- Modal-to-AdaptiveCard converter (modals.ts)
- Bridge adapter sends empty body (not "{}") for dialog close responses
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor(teams): use @microsoft/teams.cards builders for Adaptive Cards
Replace hand-rolled plain JSON objects and local type definitions with
typed builder classes from @microsoft/teams.cards. This gives compile-time
type safety and eliminates the local AdaptiveCard/AdaptiveCardElement/
AdaptiveCardAction interfaces.
Also fix ephemeral modal button missing actionType="modal", which
prevented the dialog from opening on Teams.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(teams): preserve contextId in modal update/push responses
Pass the original contextId through to re-rendered modals so subsequent
submissions can still retrieve the stored thread/message/channel context.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor: address review feedback across dialog support
- Clean up timeout timer in handleDialogOpen to prevent resource leak
- Also race on actionPromise so errors surface instead of silently timing out
- Extract buildContinueResponse helper to deduplicate update/push cases
- Use typed TextInputOptions/ChoiceSetInputOptions instead of Record<string, unknown>
- Make processSlashCommand options parameter explicit (WebhookOptions | undefined)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(chat): await storeModalContext before opening modal
The state write was fire-and-forget, so a fast dialog.submit could
arrive before the context was persisted, causing retrieveModalContext
to return empty. Also adds changeset for the new public API surface.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(chat): add missing disabled prop to ButtonProps JSX interface
The ButtonElement and ButtonOptions already supported disabled, but the
JSX ButtonProps interface was missing it, causing <Button disabled> to
silently drop the prop.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: address PR review feedback for dialog support
- Add unit tests for modals.ts (16 tests) and modal button actionType in cards.test.ts
- Make dialog open timeout configurable via dialogOpenTimeoutMs in TeamsAdapterConfig
- Fix ModalSubmitHandler type to accept Promise<void> returns, remove @ts-expect-error from example
- Delete stored modal context after retrieval to prevent state adapter leaks
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: lint fixes and add X-User-Agent header to Teams adapter
- Fix import ordering, formatting, and non-null assertions in modals.test.ts
- Sort interface members in TeamsAdapterConfig
- Add X-User-Agent: Vercel.ChatSDK header to Teams SDK App client
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(adapter-slack): replace empty table cells with single space
Slack's Block Kit API rejects cells with empty text fields. Fall back
to a single space in both mdastTableToSlackBlock (markdown.ts) and
convertTableToBlocks (cards.ts) to satisfy the API constraint.
* chore: add changeset for slack empty table cells fix
---------
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* Migrate Teams adapter from botbuilder to @microsoft/teams.apps
Replace botbuilder/botframework-connector with @microsoft/teams.apps (TeamsSDK v2.0.6).
Breaking changes:
- Remove certificate auth (TeamsAuthCertificate) — use token or managedIdentityClientId instead
- Env vars changed: TEAMS_APP_ID → CLIENT_ID, TEAMS_APP_PASSWORD → CLIENT_SECRET, TEAMS_APP_TENANT_ID → TENANT_ID
- Reactions now work (addReaction/removeReaction) instead of throwing NotImplementedError
- Graph API calls use @microsoft/teams.graph-endpoints typed endpoints
Key changes:
- BridgeHttpAdapter captures TeamsSDK route handler for serverless dispatch
- Event handlers registered via app.on() (message, messageReaction, card.action, conversationUpdate, installationUpdate)
- Outbound calls use app.send() and app.api.conversations.activities()
- Stream support via post+edit (native HttpStream when SDK exports it)
- Graph API uses typed endpoints from @microsoft/teams.graph-endpoints
- isMention detection via activity entities instead of bot name matching
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Align TeamsAdapterConfig with AppOptions from TeamsSDK
- Replace custom config fields with Pick<AppOptions, ...>
clientId (was appId), clientSecret (was appPassword), tenantId (was appTenantId)
- Remove TeamsAuthFederated, appType — use managedIdentityClientId directly
- Constructor passes config through to App (App handles env var resolution)
- Use this.app.id instead of this.config.appId for bot identity checks
- 3 replay-fetch-messages tests have known failures (type shape mismatch)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Remove api() getter, add handler type safety
- Remove useless private api() getter, use this.app.api directly
- Type handlers with specific activity types:
handleMessageActivity(ctx: IActivityContext<IMessageActivity>)
handleAdaptiveCardAction(ctx: IActivityContext<IAdaptiveCardActionInvokeActivity>)
handleReactionFromContext(ctx: IActivityContext<IMessageReactionActivity>)
- Remove unnecessary casts now that ctx.activity is properly typed
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Revert bot.tsx example changes from migration branch
Keep Azure OpenAI and DM handler changes as local-only, not part of the migration.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Revert example package.json from migration branch
Keep only adapters.ts config changes, revert package.json dep changes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Remove unrelated skill files from migration branch
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Fix
* Update README for TeamsSDK migration
- New env vars: CLIENT_ID, CLIENT_SECRET, TENANT_ID
- New config: clientId, clientSecret, tenantId, token, managedIdentityClientId
- Remove certificate auth docs (dropped)
- Reactions now supported (add/remove)
- Typing indicator now supported
- Graph API uses @microsoft/teams.graph-endpoints
- Note streaming and modals status
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Revert "Update README for TeamsSDK migration"
This reverts commit 729fe463c1.
* update readme
* Refactor adapter-teams: extract types, errors, and graph-api modules
Split the 2,114-line index.ts into focused modules:
- types.ts: TeamsAdapterConfig, TeamsThreadId, TeamsChannelContext
- errors.ts: handleTeamsError as standalone pure function
- graph-api.ts: TeamsGraphReader class with all Graph API read methods
index.ts shrinks to ~1,133 lines, delegating graph reads to TeamsGraphReader
which receives dependencies via constructor injection.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Clean up tests
* Do not break public facing config.
* Use previous .env vars
* Fix tests
* Fix webhook concurrency issue
* Remove support for reactions
* Fetch aadgroup id if it doesn't exist
* Fix lint and formatting issues in Teams adapter
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Fix README to match public API, fix 403 error mapping, correct NotImplementedError args, remove stale test aliases, and add changeset
* Fix remaining mockBotAdapter references and trailing whitespace
* fix graph pagination
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Ben Sabic <bensabic@users.noreply.github.com>
* feat(docs): add inline logo chips to hero description
Add brand logos (Slack, Microsoft Teams, Google Chat, Discord) as
inline chips prefixed before each product name in the hero section.
- Import existing SVG logos from lib/logos.tsx
- Create LogoChip component with whitespace-nowrap to prevent
wrapping between logo and its corresponding product name
- Update Hero component description type from string to ReactNode
- Keep plain text description for metadata/SEO
* fix(docs): logo chips line-height, comma wrapping, and spacing
- Use absolute positioning inside a zero-height inline-block container
so the logo doesn't increase line height
- Move trailing commas inside the whitespace-nowrap span via a suffix
prop to prevent hanging commas at line start
- Add 2px extra spacing on the right side of logos before names
* fix(docs): vertically center logos and add 2px left spacing
- Use align-middle + top-1/2 -translate-y-1/2 for true vertical
centering relative to the text line
- Add ml-[2px] on the outer nowrap span for left-side spacing
* fix(docs): widen hero description max-width on desktop
Bump from max-w-3xl (48rem) to sm:max-w-[52rem] so the first
product (Slack) stays on the first line at typical desktop widths.
* fix(docs): resolve lint issues
- Fix import order (react type import after next)
- Replace namespace import with named imports for logos
- Sort CSS classes per biome nursery rule
- Fix formatter line wrapping
---------
Co-authored-by: jamesvclements <jamesvclements@users.noreply.github.com>
* Fix Redis state typings for url and client options
* Wait for injected Redis clients to become ready
* Simplify Redis readiness guard
* Stabilize memory state TTL test with fake timers
* Handle Redis reconnect errors until client is ready
* Handle injected Redis client reconnects correctly
Ben Sabic