Issues

• Basic HTTP auth drops credentials with empty username or password

  google/adk-python#8139 · Filter by author 1aifanatic opened on Nov 13, 2026

• Incremental in-memory ingestion keeps recalling rewound turns

  Bug

  google/adk-python#8137 · Filter by author 1aifanatic opened on Nov 13, 2026
  1

• OpenAPI duplicate parameter suffixes collide with existing parameter names

  Bug

  google/adk-python#8133 · Filter by author 1aifanatic opened on Nov 13, 2026

• Evaluation pairs every ID-less tool call with the last tool response

  Bug

  google/adk-python#8132 · Filter by author 1aifanatic opened on Nov 13, 2026

• ApigeeLlm Chat Completions crashes on typed tool result values

  Bug

  google/adk-python#8131 · Filter by author 1aifanatic opened on Nov 13, 2026

• Public way to cancel an invocation without sealing it as aborted (2.14)

  Feature

  google/adk-python#8130 · Filter by author markwang-obin opened on Nov 13, 2026

• Can pin to litellm version be changed?

  google/adk-python#8129 · Filter by author Marco-Sulla opened on Nov 13, 2026

• Event compaction accepts a truncated (MAX_TOKENS) or blocked (SAFETY) summary and replaces the conversation history with it

  Bug

  google/adk-python#8127 · Filter by author HYYAlex opened on Nov 13, 2026
  1

• Typed guardrail exit-state on the confirmation ledger (PASS / DEGRADE / VETO)

  google/adk-python#8126 · Filter by author PHM-Audit opened on Nov 13, 2026

• LiteLlm sends tool results that hold a date, Decimal or Pydantic model to the model as a Python repr

  Bug

  google/adk-python#8123 · Filter by author ravismash opened on Nov 13, 2026

• MCP tool results containing ImageContent reach the model as base64 text instead of image parts

  Bug

  google/adk-python#8118 · Filter by author omnomkar opened on Nov 13, 2026
  1

  

• OpenAIResponsesLlm should declare output_schema_and_tools on the real OpenAI host

  Feature

  google/adk-python#8115 · Filter by author nicolasmota opened on Nov 12, 2026
  

• OpenAIResponsesLlm drops reasoning items between tool calls (never replayed or chained)

  Bug

  google/adk-python#8110 · Filter by author QueryPlanner opened on Nov 12, 2026
  3

  

• [LiteLLM] Non-Anthropic reasoning_details (encrypted/signed) are dropped on replay

  Bug

  google/adk-python#8109 · Filter by author QueryPlanner opened on Nov 12, 2026
  4

  

• feat(skills): let a classifier load the skill for a turn before the model's first step

  needs reviewtools
  Feature

  google/adk-python#8107 · Filter by author lwangverizon opened on Nov 12, 2026
  1

  

• Repeated cancellation can interrupt root cleanup and leave HTTP connections in use

  Bug

  google/adk-python#8106 · Filter by author auwilly opened on Nov 12, 2026
  4

  

• Opt-in way to stop sliding-window compaction from delaying the end of Runner.run_async (and of /run_sse)

  Feature

  google/adk-python#8105 · Filter by author hugueslys opened on Nov 12, 2026
  

• displays Ollama/Qwen3 thought text as part of the final answer

  modelsrequest clarification
  Bug

  google/adk-python#8102 · Filter by author lahiruramesh opened on Nov 12, 2026
  1

  

• ApigeeLlm (Chat Completions) ignores FunctionCallingConfig.allowed_function_names and the VALIDATED mode

  Bug

  google/adk-python#8100 · Filter by author Android-Login-Analysis opened on Nov 11, 2026
  2

  

• Allow openai 3.x in the openai extra (currently openai>=2.20,<3)

  modelsneeds review
  Feature

  google/adk-python#8099 · Filter by author rcleveng opened on Nov 11, 2026
  2

  

• Proposal: add a self-healing SQL analyst sample (in-memory SQLite + LoopAgent)

  Feature

  google/adk-python#8098 · Filter by author themonusharma opened on Nov 11, 2026
  1

  

• GenAI telemetry exports thought parts as text instead of reasoning

  Bug

  google/adk-python#8095 · Filter by author juriwiens opened on Nov 11, 2026
  3

  

• Workflow skips graph validation and drops its output when given a graph directly

  Bug

  google/adk-python#8090 · Filter by author hyeonsang010716 opened on Nov 11, 2026
  1

  

• Run evaluation fails with "ValueError: No parts in the new_message." after creating fresh test case

  evalrequest clarification
  Bug

  google/adk-python#8080 · Filter by author yuwtennis opened on Nov 10, 2026
  5

  

• Proposal: deterministic tool-call expectations for conversation_scenario eval cases

  evalneeds review
  Feature

  google/adk-python#8077 · Filter by author Fruiterer333 opened on Nov 10, 2026
  3