Changelog

New Features:

• Superserve Tools: Added SuperserveTools to run agent-generated code and manage files in Superserve, a Firecracker-based sandbox platform for long-running agents. See docs.
• Plivo Tools: Added PlivoTools to send SMS, make voice calls, and look up phone numbers with Plivo. See docs.
• The Context Company: Added an observability integration to trace agent runs with The Context Company. See docs

Improvements:

• agno create Enhancements: Interactive starter template and project name prompts, four new starters (Azure, Helm, Modal, Render), and automatic .env seeding.
• Telegram Tools: Add pin_message, get_chat, get_file, and react_with_emoji, plus save_downloads and output_directory to save downloaded files to disk.
• Tavily Tools: Add domain, date range, topic, and country filters to searches.
• Google Tools: Add pagination to Gmail, Calendar, Drive, and Slides.
• Oxylabs Tools: Return full page content as Markdown.
• Workflows: Accept run_context in Router selectors and Condition evaluators, deprecating session_state.

Bug Fixes:

• Teams:
  • Deliver user input to members that pause more than once in a run.
  • Serialize run outputs whose media or member responses are already dicts.
• AgentOS:
  • Skip accumulated run outputs when streaming team responses.
  • Reject creating a session with an existing id instead of overwriting its history.
• Workflow & Team Sessions: Honor zero and negative history limits.
• Workflows: Surface the underlying error when a workflow fails mid-stream.
• Slack: Scope sessions per channel so channels no longer share history.
• Tavily Tools: Apply the configured extract depth.
• OpenRouter: Return generated images from image models.
• PostgresDb: Respect create_schema=False when components override table names.
• PubMed, Firecrawl & Spider Tools: Respect configured and per-call result limits.
• FixedSizeChunking: Keep short documents when the overlap exceeds the content.
• CSVReader & ExcelReader: Give each reader its own default row chunking.

What's Changed

• fix: CSVReader/ExcelReader no longer share one mutable default RowChunking instance by @chuenchen309 in #8922
• fix: FixedSizeChunking silently drops short documents when overlap >= content length by @chuenchen309 in #8943
• [fix] make POST /sessions idempotent for an existing session_id by @ormeilu in #8646
• [fix] Skip run output accumulators in team SSE streams by @cyphercodes in #8236
• fix: guard zero/negative run and message counts in WorkflowSession/TeamSession helpers by @chuenchen309 in #8942
• feat: Add PlivoTools toolkit for SMS and Voice by @sarveshpatil-plivo in #8868
• fix: return generated images from OpenRouter responses by @Himanshu040604 in #8938
• fix: respect configured and explicit result limits in PubMed, Firecrawl and Spider tools by @Orion7r in #6401
• fix: handle already-serialized dicts in TeamRunOutput.to_dict media fields and member_responses by @kausmeows in #8966
• feat: add react_with_emoji tool to TelegramTools by @flobo3 in #7369
• fix: remove deprecated show_tool_calls from cookbooks by @Mustafa-Esoofally in #8897
• fix: channel-scoped Slack session keys to prevent cross-channel collision by @Mustafa-Esoofally in #8933
• cookbook: add The Context Company observability example by @rohilvagarwal in #8955
• fix: preserve Postgres create_schema in resolve_db_from_config clone path by @shaun0927 in #7547
• chore: remove shared default OpenAI model util and refresh example model ids by @maxpetrusenkoagent in #8384
• feat: add pin_message, get_chat, get_file to TelegramTools by @shaunymca in #6548
• feat: improve agno create onboarding by @ashpreetbedi in #8959
• feat: add pagination and max_results_per_request to Gmail tools by @Mustafa-Esoofally in #8462
• feat: expose advanced search parameters in tavily tools by @Himanshu040604 in #8896
• chore: deprecate session_state in condition/router workflow in favour of run_context by @kausmeows in #9018
• fix: persist paused member runs for multi-round team HITL by @Mustafa-Esoofally in #8937
• feat: Add Superserve sandbox tools by @mohamedsobhi777 in #8584
• chore: parallelize CI tests with pytest-split by @Mustafa-Esoofally in #8831
• fix: update oxylabs scrape_website method by @oxy-giedrius in #4882
• fix: Revert "feat: add pagination and max_results_per_request to Gmail tools" by @kausmeows in #9025
• chore: Release v2.7.4 by @harshsinha03 in #9019

New Contributors

• @cyphercodes made their first contribution in #8236
• @sarveshpatil-plivo made their first contribution in #8868
• @Orion7r made their first contribution in #6401
• @rohilvagarwal made their first contribution in #8955
• @maxpetrusenkoagent made their first contribution in #8384
• @shaunymca made their first contribution in #6548
• @mohamedsobhi777 made their first contribution in #8584

Full Changelog: v2.7.3...v2.7.4