Feature/social media integration testing by pulinduvidmal · Pull Request #233 · yaalalabs/agent-kernel

pulinduvidmal · 2026-02-16T18:44:43Z

Description

Add real Slack/Telegram integration test

Type of Change

Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to not work as expected)
Documentation update
Refactoring (no functional changes)
Performance improvement
Test update
CI/CD update
Other (please describe):

Related Issues

Fixes #
Relates to #

Changes Made

Testing

Unit tests pass locally
Integration tests pass locally
Manual testing completed
New tests added for changes

Checklist

My code follows the project's style guidelines
I have performed a self-review of my code
I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
My changes generate no new warnings
I have added tests that prove my fix is effective or that my feature works
New and existing unit tests pass locally with my changes
Any dependent changes have been merged and published

Screenshots (if applicable)

Additional Notes

…upport_for_telegram

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

…support

Copilot

Pull request overview

Copilot reviewed 41 out of 47 changed files in this pull request and generated 8 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-02T11:10:06Z

ak-py/src/agentkernel/core/config.py

+    max_attachments: int = Field(default=20, description="Maximum number of attachments to keep per session")
+    description_max_length: int = Field(default=200, description="Maximum length of attachment description text")
+    description_model: str = Field(
+        default="gpt-4o",
+        description="LiteLLM model used to generate brief descriptions when an attachment is first received (called by the pre-hook)",
+    )
+    analysis_model: str = Field(
+        default="gpt-4o",
+        description="LiteLLM model used by the analyze_attachments tool when the agent requests a full analysis of an attachment",
+    )


description_max_length is introduced in _MultimodalConfig, but it isn’t referenced anywhere else in the codebase. Either enforce this limit when generating/storing descriptions (e.g., in describe_attachment_briefly or the multimodal pre-hook) or remove the config option to avoid dead/unused configuration knobs.

Copilot · 2026-03-02T11:10:06Z

examples/api/telegram/server_test.py

 import asyncio
+import json
 import os


json is imported but never used in this test module. Remove it to keep the test file clean (and to avoid unused-import lint failures if enabled).

Copilot · 2026-03-02T11:10:07Z

docs/docs/integrations/teams.md

+3.  Click **Apply**.
+4.  Go to **Channels** and add **Microsoft Teams**. accepting the terms.
+


Minor grammar: sentence is split oddly here ("Go to Channels and add Microsoft Teams. accepting the terms."). Consider changing to a single sentence like "...add Microsoft Teams, accepting the terms." or "...add Microsoft Teams and accept the terms.".

Copilot · 2026-03-02T11:10:07Z

ak-py/src/agentkernel/integration/slack/slack_chat.py

        self._log = logging.getLogger("ak.api.slack")
        self._slack_agent = Config.get().slack.agent if Config.get().slack.agent != "" else None
        self._slack_agent_acknowledgement = Config.get().slack.agent_acknowledgement if Config.get().slack.agent_acknowledgement != "" else None
+        self._slack_test_mode = Config.get().slack.test_mode


Config.get().slack.test_mode is accessed here, but _SlackConfig in ak-py/src/agentkernel/core/config.py does not define a test_mode field. With extra="ignore", test_mode in YAML/env will be discarded and this will raise AttributeError at runtime during handler initialization. Add test_mode: bool = Field(default=False, ...) to _SlackConfig (and document it), or use getattr(Config.get().slack, "test_mode", False) as a backwards-compatible default.

Suggested change

self._slack_test_mode = Config.get().slack.test_mode

self._slack_test_mode = getattr(Config.get().slack, "test_mode", False)

Copilot · 2026-03-02T11:10:07Z

ak-py/src/agentkernel/integration/telegram/telegram_chat.py

        self._telegram_agent = Config.get().telegram.agent if Config.get().telegram.agent != "" else None
        self._bot_token = Config.get().telegram.bot_token
        self._webhook_secret = Config.get().telegram.webhook_secret
        self._api_version = Config.get().telegram.api_version or "bot"
-        self._base_url = f"https://api.telegram.org/{self._api_version}{self._bot_token}"
+        self._api_base_url = Config.get().telegram.api_base_url
+        self._base_url = f"{self._api_base_url}/{self._api_version}{self._bot_token}"



Config.get().telegram.api_base_url is used to build _base_url, but _TelegramConfig in ak-py/src/agentkernel/core/config.py does not define api_base_url. This will raise AttributeError at startup, and env vars like AK_TELEGRAM__API_BASE_URL will be ignored due to extra="ignore". Add api_base_url to _TelegramConfig with a default like https://api.telegram.org (and consider normalizing trailing / to avoid double slashes).

Copilot · 2026-03-02T11:10:07Z

ak-py/src/agentkernel/integration/teams/teams_chat.py

+                # Bot Framework expects a dict with 'body' as parsed JSON
+                req_dict = {"body": json.loads(body_text), "headers": {"Authorization": auth_header}}  # Parse JSON
+
+                # Process the activity using Bot Framework adapter
+                async def bot_logic(turn_context: TurnContext):
+                    await self._handle_teams_message(turn_context)
+
+                # Process the request - Bot Framework validates auth internally
+                await self._adapter.process_activity(req_dict, auth_header, bot_logic)


BotFrameworkAdapter.process_activity(...) expects an Activity (deserialized from the request JSON), not a custom {"body": ..., "headers": ...} dict. As written, this is very likely to fail inside the adapter (and you’re also discarding any InvokeResponse that should be returned for certain activity types). Deserialize with Activity().deserialize(json.loads(body_text)), call process_activity(activity, auth_header, bot_logic), and if an invoke response is returned, convert it into the FastAPI Response (status/body/headers).

Suggested change

# Bot Framework expects a dict with 'body' as parsed JSON

req_dict = {"body": json.loads(body_text), "headers": {"Authorization": auth_header}} # Parse JSON

# Process the activity using Bot Framework adapter

async def bot_logic(turn_context: TurnContext):

await self._handle_teams_message(turn_context)

# Process the request - Bot Framework validates auth internally

await self._adapter.process_activity(req_dict, auth_header, bot_logic)

# Deserialize the incoming request body into a Bot Framework Activity

activity = Activity().deserialize(json.loads(body_text))

# Process the activity using Bot Framework adapter

async def bot_logic(turn_context: TurnContext):

await self._handle_teams_message(turn_context)

# Process the request - Bot Framework validates auth internally

invoke_response = await self._adapter.process_activity(activity, auth_header, bot_logic)

# If an InvokeResponse is returned, convert it into a FastAPI Response

if invoke_response:

body = invoke_response.body if hasattr(invoke_response, "body") else None

content = json.dumps(body) if body is not None else None

headers = getattr(invoke_response, "headers", None)

if headers is not None and not isinstance(headers, dict):

headers = None

return Response(

status_code=invoke_response.status,

content=content,

media_type="application/json" if content is not None else None,

headers=headers,

)

# For non-invoke activities, return a simple 200 OK

Copilot · 2026-03-02T11:10:08Z

examples/api/slack/server_test.py

+Usage:
+    cd examples/api/slack && ./build.sh local
+    uv run pytest server_integration_test.py -s --junitxml=pytest-report.xml
+"""


The usage string references server_integration_test.py, but this file is server_test.py (and there’s no server_integration_test.py in the repo). Update the docstring so the command matches the actual filename to avoid confusing users/CI scripts.

Copilot · 2026-03-02T11:10:08Z

examples/api/slack/server_test.py

+async def _read_channel_messages() -> list[dict]:
+    """Read the most recent message from the dedicated test channel."""
+    async with httpx.AsyncClient(timeout=10.0) as client:
+        resp = await client.get(
+            "https://slack.com/api/conversations.history",
+            params={
+                "channel": CHANNEL_ID,
+                "limit": 1,
+            },
+            headers={"Authorization": f"Bearer {BOT_TOKEN}"},
+        )
+        resp.raise_for_status()
+        data = resp.json()
+
+        print(
+            f"conversations.history response: ok={data.get('ok')}, "
+            f"messages={len(data.get('messages', []))}, "
+            f"error={data.get('error', 'none')}"
+        )
+
+        if not data.get("ok"):
+            raise RuntimeError(f"Slack API error: {data.get('error', 'unknown')}")
+
+        return data.get("messages", [])


conversations.history is queried with limit: 1, and the test assumes the most recent channel message is the acknowledgment from this run. If any other message is posted to the channel (human, another test, Slack system message), the test can read/update/delete the wrong message and become flaky. Consider fetching a larger window (e.g., 10–50 messages) and selecting the message by a unique marker (uuid/run id) included in the webhook/ack text, or by timestamp >= test start time.

Copilot

Pull request overview

Copilot reviewed 43 out of 49 changed files in this pull request and generated 12 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-02T18:01:32Z

ak-py/src/agentkernel/integration/teams/teams_chat.py

+                failed_files = await self._process_attachments(activity.attachments, requests)
+                if failed_files:
+                    await turn_context.send_activity(
+                        f"Sorry {user_name}, I could not download the following files: {', '.join(failed_files)}. Please try again."


The block returns failed_files + rejected_files + oversized_files, but the user-facing message in _handle_teams_message says they "could not download" all of them. Rejected (audio/video) and oversized files aren’t download failures, so the error text should distinguish these cases (or return structured categories instead of a single list).

Suggested change

f"Sorry {user_name}, I could not download the following files: {', '.join(failed_files)}. Please try again."

f"Sorry {user_name}, I could not use the following files (they may be unsupported types, too large, or failed to download): {', '.join(failed_files)}. Please try again."

Copilot · 2026-03-02T18:01:32Z

ak-py/src/agentkernel/integration/teams/teams_chat.py

@@ -0,0 +1,301 @@
+import base64
+import io


io is imported but never used in this module.

Suggested change

import io

Copilot · 2026-03-02T18:01:32Z

examples/api/slack/server_test.py

+Usage:
+    cd examples/api/slack && ./build.sh local
+    uv run pytest server_integration_test.py -s --junitxml=pytest-report.xml
+"""


The usage instructions reference server_integration_test.py, but this repository only contains server_test.py. This makes the copy/paste instructions incorrect.

Copilot · 2026-03-02T18:01:33Z

examples/api/telegram/server_test.py

 import asyncio
+import json
 import os


json is imported but never used (the code calls request.json(), which doesn’t require the stdlib json module). Please remove the unused import to avoid lint/test failures in stricter environments.

Copilot · 2026-03-02T18:01:33Z

examples/api/whatsapp/server_test.py

 import asyncio
+import json
 import os


json is imported but never used (the code calls request.json(), which doesn’t require the stdlib json module). Please remove the unused import to avoid lint/test failures in stricter environments.

Copilot · 2026-03-02T18:01:34Z

ak-py/src/agentkernel/integration/whatsapp/whatsapp_chat.py

+        import os
+
+        base_url_env = os.environ.get("AK_WHATSAPP__API_BASE_URL", "").strip()
+        if not base_url_env:
+            base_url_env = Config.get().whatsapp.api_base_url
+        base_url = f"{base_url_env}/{self._api_version}"


self._base_url is computed in __init__ from AK_WHATSAPP__API_BASE_URL but _send_message() recomputes a separate base_url (and strips whitespace) instead of using self._base_url. This can lead to inconsistent behavior where media downloads / mark-as-read use self._base_url while sending messages uses the recomputed URL (especially if the env var has whitespace or a trailing slash). Consider normalizing the base URL once in __init__ (strip + rstrip('/')) and reusing it everywhere.

Suggested change

import os

base_url_env = os.environ.get("AK_WHATSAPP__API_BASE_URL", "").strip()

if not base_url_env:

base_url_env = Config.get().whatsapp.api_base_url

base_url = f"{base_url_env}/{self._api_version}"

# Use the base URL configured on initialization to ensure consistency

base_url = f"{self._base_url.strip().rstrip('/')}/{self._api_version}"

Copilot · 2026-03-02T18:01:34Z

docs/docs/integrations/telegram.md

+```bash
+export AK_MULTIMODAL__ENABLED=true              # Enable multimodal support (default: false)
+export AK_MULTIMODAL__MAX_ATTACHMENTS=5         # Keep last N files in session (default: 5)
+export AK_MULTIMODAL__ATTACHMENT_TTL=604800     # File lifetime in seconds (default: 604800 = 1 week)
+```


The docs list AK_MULTIMODAL__ATTACHMENT_TTL and state defaults for AK_MULTIMODAL__MAX_ATTACHMENTS, but the current config model only defines multimodal.max_attachments (default 20) and there is no TTL setting/implementation in the code. Either implement TTL + align defaults, or update this block to reflect the actual supported configuration.

Copilot · 2026-03-02T18:01:34Z

examples/api/telegram/README.md

+```bash
+export AK_MULTIMODAL__ENABLED=true              # Enable multimodal support (default: false)
+export AK_MULTIMODAL__MAX_ATTACHMENTS=5         # Keep last N files in session (default: 5)
+export AK_MULTIMODAL__ATTACHMENT_TTL=604800     # File lifetime in seconds (default: 604800 = 1 week)
+```


This README documents AK_MULTIMODAL__ATTACHMENT_TTL and claims AK_MULTIMODAL__MAX_ATTACHMENTS defaults to 5, but the current Agent Kernel config only supports multimodal.max_attachments (default 20) and has no TTL setting. Please update the README to match the implemented config, or add the missing config fields + TTL behavior.

Copilot · 2026-03-02T18:01:34Z

docs/docs/integrations/teams.md

+2.  **Messaging endpoint**: Enter your server's public URL + `/teams/messages`.
+    *   Example: `https://your-server.com/teams/messages`
+3.  Click **Apply**.
+4.  Go to **Channels** and add **Microsoft Teams**. accepting the terms.


Grammar: "Go to Channels and add Microsoft Teams. accepting the terms." should be a complete sentence (e.g., "..., accepting the terms" or "... and add ..., accepting the terms").

Copilot · 2026-03-02T18:01:34Z

ak-py/src/agentkernel/core/config.py

 class _APIConfig(BaseModel):
    host: str = Field(default="0.0.0.0", description="API host")
    port: int = Field(default=8000, description="API port")
    enabled_routes: _RoutesConfig = Field(description="API route flags", default_factory=_RoutesConfig)
    custom_router_prefix: str = Field(default="/custom", description="Custom router prefix")
-    max_file_size: int = Field(default=2097152, description="Maximum file size in bytes (default: 2 MB)")
+    max_file_size: int = Field(default=20971520, description="Maximum file size in bytes (default: 20 MB)")


 class _A2AConfig(BaseModel):
    enabled: bool = Field(default=False, description="Enable A2A")
    agents: List[str] = Field(default=["*"], description="List of agent names to enable A2A")
    url: str = Field(default="http://localhost:8000/a2a", description="A2A URL")
    task_store_type: str = Field(default="in_memory", pattern="^(in_memory|redis)$")


 class _MCPConfig(BaseModel):
    enabled: bool = Field(default=False, description="Enable MCP")
    expose_agents: bool = Field(default=False, description="Expose agents as MCP tools")
    agents: List[str] = Field(default=["*"], description="List of agent names to expose as MCP tool")
    url: str = Field(default="http://localhost:8000/mcp", description="MCP URL")


 class _SlackConfig(BaseModel):
    agent: str = Field(default="", description="Default agent to use for Slack interactions")
    agent_acknowledgement: str = Field(
        default="",
        description="The message to send as an acknowledgement when a Slack message is received",
    )
+    test_mode: bool = Field(default=False, description="Enable test mode for Slack integration")
+
+
+class _TeamsConfig(BaseModel):
+    agent: str = Field(default="", description="Default agent to use for Microsoft Teams interactions")
+    agent_acknowledgement: str = Field(
+        default="",
+        description="The message to send as an acknowledgement when a Teams message is received",
+    )
+    app_id: str = Field(default="", description="Microsoft App ID (Application/Client ID) from Azure AD App Registration")
+    app_password: str = Field(default="", description="Microsoft App Password (Client Secret) from Azure AD App Registration")
+    tenant_id: str = Field(default="", description="Optional Tenant ID for Single Tenant App registration")


PR description frames this as a Slack/Telegram integration test/refactor, but this change set also introduces new public config surface (teams, multimodal, slack.test_mode) and changes defaults like api.max_file_size (2MB -> 20MB). Please update the PR description/type-of-change checklist to reflect these functional additions, since they impact library behavior beyond tests.

Copilot

Pull request overview

Copilot reviewed 47 out of 54 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-02T19:13:13Z

ak-py/src/agentkernel/integration/whatsapp/whatsapp_chat.py

+        import os
+
+        base_url_env = os.getenv("AK_WHATSAPP__API_BASE_URL")
+        self._base_url = f"{base_url_env or Config.get().whatsapp.api_base_url}/{self._api_version}"


AK_WHATSAPP__API_BASE_URL is being read directly via os.getenv() inside __init__, even though api_base_url is now a first-class config field (Config.get().whatsapp.api_base_url) and will already be populated from the same env var by the settings layer. This duplication also makes precedence/whitespace handling inconsistent (here the env var isn’t .strip()ed). Consider relying on Config.get().whatsapp.api_base_url only (and normalizing it once) and removing the inline import os.

Copilot · 2026-03-02T19:13:13Z

ak-py/src/agentkernel/integration/whatsapp/whatsapp_chat.py

+        import os
+
+        base_url_env = os.environ.get("AK_WHATSAPP__API_BASE_URL", "").strip()
+        if not base_url_env:
+            base_url_env = Config.get().whatsapp.api_base_url
+        base_url = f"{base_url_env}/{self._api_version}"
+        url = f"{base_url}/{self._phone_number_id}/messages"



_send_message() re-reads AK_WHATSAPP__API_BASE_URL and recomputes the base URL even though self._base_url was already constructed in __init__. This duplication makes it easy for the handler to use different base URLs in different code paths (and it adds another inline import os). Consider computing/normalizing the base URL once (in config or __init__) and reusing that value here.

Suggested change

import os

base_url_env = os.environ.get("AK_WHATSAPP__API_BASE_URL", "").strip()

if not base_url_env:

base_url_env = Config.get().whatsapp.api_base_url

base_url = f"{base_url_env}/{self._api_version}"

url = f"{base_url}/{self._phone_number_id}/messages"

url = f"{self._base_url}/{self._phone_number_id}/messages"

Copilot · 2026-03-02T19:13:14Z

ak-py/src/agentkernel/core/runtime.py

+    _system_pre_hooks: list = [MultimodalPreHookFactory.get(), InputGuardrailFactory.get()]
    _system_post_hooks: list = [OutputGuardrailFactory.get()]


_system_pre_hooks is initialized at import/class-definition time via MultimodalPreHookFactory.get(). If multimodal is enabled/disabled via env/config after import (or in tests that monkeypatch config), this list will keep the originally-created hook (often the NoOp hook), so multimodal pre-processing may never run. Consider instantiating the multimodal hook inside Runtime.__init__ (or always registering MultimodalPreHook and letting it check config in on_run) instead of freezing it as a class-level value.

Copilot

Pull request overview

Copilot reviewed 49 out of 56 changed files in this pull request and generated 5 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-03T11:23:13Z

examples/api/gmail/server_test.py

+    service = get_gmail_service()
+    if not service:
+        pytest.skip(f"Valid token.pickle not found at {TOKEN_FILE}. Cannot run real Gmail test.")
+


get_gmail_service() can return None for reasons other than a missing token.pickle (e.g., missing AK_GMAIL__CLIENT_ID/AK_GMAIL__CLIENT_SECRET), but the skip reason here always claims the token file is missing. Consider returning a more specific error/enum from get_gmail_service() or updating the skip message to cover all None cases.

Copilot · 2026-03-03T11:23:14Z

ak-py/src/agentkernel/integration/teams/teams_chat.py

@@ -0,0 +1,301 @@
+import base64
+import io


io is imported but never used in this module. Please remove the unused import to avoid lint/test failures in stricter environments.

Suggested change

import io

Copilot · 2026-03-03T11:23:14Z

docs/docs/integrations/teams.md

+2.  **Messaging endpoint**: Enter your server's public URL + `/teams/messages`.
+    *   Example: `https://your-server.com/teams/messages`
+3.  Click **Apply**.
+4.  Go to **Channels** and add **Microsoft Teams**. accepting the terms.


Grammar: "add Microsoft Teams. accepting the terms." should be a complete sentence (e.g., "...and add Microsoft Teams, accepting the terms.").

Copilot · 2026-03-03T11:23:14Z

examples/api/whatsapp/server_test.py

 import asyncio
+import json
 import os


json is imported but not used in this test module. Please remove the unused import to keep the integration test tidy.

Copilot · 2026-03-03T11:23:14Z

ak-py/src/agentkernel/core/base.py

+        if config and config.enabled:
+            self.override_system_prompt(session=None, prompt=self.get_system_prompt_suffix())


override_system_prompt is declared to require a Session, but _setup_system_prompt() calls it with session=None. This is inconsistent with the abstract method contract and may break implementations that assume a real session. Consider typing the parameter as Session | None (and updating implementations) or refactor init-time prompt injection to not require a session object.

Suggested change

if config and config.enabled:

self.override_system_prompt(session=None, prompt=self.get_system_prompt_suffix())

session = getattr(self, "session", None)

if config and config.enabled and session is not None:

self.override_system_prompt(session=session, prompt=self.get_system_prompt_suffix())

Copilot

Pull request overview

Copilot reviewed 50 out of 57 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-03T12:50:33Z

ak-py/src/agentkernel/integration/teams/teams_chat.py

+                    await turn_context.send_activity(
+                        f"Sorry {user_name}, I could not download the following files: {', '.join(failed_files)}. Please try again."
+                    )
+                    return


_process_attachments() returns a combined list of (download failures + rejected audio/video + oversized files), but the caller treats this purely as “could not download” and aborts the whole request. Consider returning a structured result (e.g., separate lists for failed/rejected/oversized) so you can send an accurate user message (and potentially still proceed when only some attachments are rejected).

Suggested change

await turn_context.send_activity(

f"Sorry {user_name}, I could not download the following files: {', '.join(failed_files)}. Please try again."

)

return

# Some attachments could not be used (e.g., failed download, rejected type, or oversized),

# but we can still proceed with the rest of the request.

await turn_context.send_activity(

f"Sorry {user_name}, I could not process the following files and will continue without them: {', '.join(failed_files)}."

)

pulinduvidmal and others added 30 commits December 23, 2025 21:13

feat: add file support for messenger integration

9014f6b

feat: add multimodal file/image support for telegram

5222577

feat: add multimodal file/image support for telegram

54def41

fix: resolve messenger integration conflicts from rebase

d0199b6

docs: update doc

698747d

fix: resolve messenger integration conflicts from rebase

bb4c1b3

feat: Add multimodal memory support

a906b91

feat: Add multimodal memory support

1cc6d53

Merge remote-tracking branch 'origin/develop' into feature/add_file_s…

5cb693f

…upport_for_telegram

fix: Add multimodal memory support

d99ae92

fix: Add multimodal memory support

d1f68e5

fix: telegram_chat.py

7c6d40c

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

docs: Add multimodal memory support

84330c1

fix telegram_chat.py

7e08a7c

fix telegram_chat.py

d9e601b

fix telegram_chat.py

8f7733c

feat: add teams integration

23d7c61

feat: add teams integration

930ed98

feat: add teams integration

c0f5652

feat: add file/image attachment support with two-step LLM

94a350f

feat: add file/image attachment support with two-step LLM

94cf7a9

refactor: split multimodal into package

e6d9ba1

refactor: split multimodal into package

b529b66

feat: add file/image attachment support with two-step LLM

28e26ba

feat: add file/image attachment support with two-step LLM

fd36019

feat(multimodal): add tools integration for attachment support

06bfb32

feat: add tools integration for attachment support

acfeba4

feat: add tools integration for attachment support

c5c2303

feat: update tenant id

b0a52ec

docs: update documents for the attachment tool

216d5a7

pulinduvidmal added 5 commits February 26, 2026 00:57

Merge add_file_support_for_telegram: Integrate Teams with multimodal …

9933d5e

…support

Merge add_file_support_for_telegram: Integrate Teams with multimodal …

3770ded

…support

chore: remove garbage file from repository

aa895b6

Merge branch 'feature/add_teams_integration'

b1cd302

Restore multimodal changes from Teams branch

74cfc87

Copilot AI review requested due to automatic review settings March 2, 2026 11:02

Copilot started reviewing on behalf of pulinduvidmal March 2, 2026 11:03 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

pulinduvidmal added 2 commits March 2, 2026 20:24

fix: config

8f296f4

feat: add whatsapp real test

fa78f65

Copilot AI review requested due to automatic review settings March 2, 2026 17:50

Copilot started reviewing on behalf of pulinduvidmal March 2, 2026 17:51 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

pulinduvidmal added 2 commits March 3, 2026 00:30

feat: add gmail test

46873bc

feat: add gmail test

ecda8f2

Copilot AI review requested due to automatic review settings March 2, 2026 19:03

Copilot started reviewing on behalf of pulinduvidmal March 2, 2026 19:03 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

pulinduvidmal added 3 commits March 3, 2026 14:53

feat: add messenger test

d97475f

feat: add messenger test

bf6c998

feat: add instagram test

1260d02

Copilot AI review requested due to automatic review settings March 3, 2026 11:14

Copilot started reviewing on behalf of pulinduvidmal March 3, 2026 11:14 View session

Copilot AI reviewed Mar 3, 2026

View reviewed changes

pulinduvidmal added 3 commits March 3, 2026 18:00

feat: add Teams integration test

8262056

feat: add Teams integration test

caa21a8

feat: add tests to nightly integration tests

6064728

Copilot AI review requested due to automatic review settings March 3, 2026 12:43

Copilot started reviewing on behalf of pulinduvidmal March 3, 2026 12:43 View session

Copilot AI reviewed Mar 3, 2026

View reviewed changes

		3. Click Apply.
		4. Go to Channels and add Microsoft Teams. accepting the terms.

	self._slack_test_mode = Config.get().slack.test_mode
	self._slack_test_mode = getattr(Config.get().slack, "test_mode", False)

-                # Bot Framework expects a dict with 'body' as parsed JSON
-                req_dict = {"body": json.loads(body_text), "headers": {"Authorization": auth_header}}  # Parse JSON
-                # Process the activity using Bot Framework adapter
-                async def bot_logic(turn_context: TurnContext):
-                    await self._handle_teams_message(turn_context)
-                # Process the request - Bot Framework validates auth internally
-                await self._adapter.process_activity(req_dict, auth_header, bot_logic)
+                # Deserialize the incoming request body into a Bot Framework Activity
+                activity = Activity().deserialize(json.loads(body_text))
+                # Process the activity using Bot Framework adapter
+                async def bot_logic(turn_context: TurnContext):
+                    await self._handle_teams_message(turn_context)
+                # Process the request - Bot Framework validates auth internally
+                invoke_response = await self._adapter.process_activity(activity, auth_header, bot_logic)
+                # If an InvokeResponse is returned, convert it into a FastAPI Response
+                if invoke_response:
+                    body = invoke_response.body if hasattr(invoke_response, "body") else None
+                    content = json.dumps(body) if body is not None else None
+                    headers = getattr(invoke_response, "headers", None)
+                    if headers is not None and not isinstance(headers, dict):
+                        headers = None
+                    return Response(
+                        status_code=invoke_response.status,
+                        content=content,
+                        media_type="application/json" if content is not None else None,
+                        headers=headers,
+                    )
+                # For non-invoke activities, return a simple 200 OK

	f"Sorry {user_name}, I could not download the following files: {', '.join(failed_files)}. Please try again."
	f"Sorry {user_name}, I could not use the following files (they may be unsupported types, too large, or failed to download): {', '.join(failed_files)}. Please try again."

		_system_pre_hooks: list = [MultimodalPreHookFactory.get(), InputGuardrailFactory.get()]
		_system_post_hooks: list = [OutputGuardrailFactory.get()]

		if config and config.enabled:
		self.override_system_prompt(session=None, prompt=self.get_system_prompt_suffix())

-        if config and config.enabled:
-            self.override_system_prompt(session=None, prompt=self.get_system_prompt_suffix())
+        session = getattr(self, "session", None)
+        if config and config.enabled and session is not None:
+            self.override_system_prompt(session=session, prompt=self.get_system_prompt_suffix())

Conversation

pulinduvidmal commented Feb 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Type of Change

Related Issues

Changes Made

Testing

Checklist

Screenshots (if applicable)

Additional Notes

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

pulinduvidmal commented Feb 16, 2026 •

edited

Loading