- Fix assertion in streaming dict fallback test (trailing space in data
not reflected in expected value).
- Add two regression tests proving that models with reasoning_content
(e.g. DeepSeek-R1) and standard models (no reasoning fields) are
completely unaffected by the reasoning fallback.
Made-with: Cursor
- Add reasoning_content fallback from reasoning in _parse dict branch
- Add content fallback from msg.reasoning in _parse SDK object branch
- Add reasoning_content fallback in _parse SDK object branch
- Add reasoning fallback in _parse_chunks dict branch
- Add reasoning fallback in _parse_chunks SDK object branch
This ensures StepFun Plan API works correctly in both streaming and
non-streaming modes, for both dict and SDK object response formats.
Add comprehensive tests for the StepFun Plan API compatibility fix:
- _parse dict branch: content and reasoning_content fallback to reasoning
- _parse SDK object branch: same fallback for pydantic response objects
- _parse_chunks dict branch: reasoning field handled in streaming mode
- _parse_chunks SDK branch: reasoning fallback for SDK delta objects
- Precedence tests: reasoning_content field takes priority over reasoning
Refs: fix(provider): support StepFun Plan API reasoning field fallback
StepFun Plan API returns response content in the 'reasoning' field when
the model is in thinking mode and 'content' is empty. OpenAICompatProvider
previously only checked 'content' and 'reasoning_content', missing this field.
This patch adds a fallback: if content is empty and 'reasoning' is present,
extract text from reasoning to populate content, ensuring StepFun models
(step-3.5-flash, step-3.5-flash-2603) work correctly with tool calls.
Co-authored-by: moranfong <moranfong@gmail.com>
- Drop *.pycs, *.pywz, *.pyzz — not real Python file extensions.
- Move nano.*.save from "Project-specific" to "Editors & IDEs" where
it belongs (nano editor backup files, not project artifacts).
Made-with: Cursor
- Propagate `description` from MCP prompt arguments into the JSON
Schema so LLMs can better understand prompt parameters.
- Align generic-exception error message with tool/resource wrappers
(drop redundant `{exc}` detail).
- Extend test fixture to mock `mcp.shared.exceptions.McpError`.
- Add tests for argument description forwarding and McpError handling.
Made-with: Cursor
Add MCPResourceWrapper and MCPPromptWrapper classes that expose MCP
server resources and prompts as nanobot tools. Resources are read-only
tools that fetch content by URI, and prompts are read-only tools that
return filled prompt templates with optional arguments.
- MCPResourceWrapper: reads resource content (text and binary) via URI
- MCPPromptWrapper: gets prompt templates with typed arguments
- Both handle timeouts, cancellation, and MCP SDK 1.x error types
- Resources and prompts are registered during server connection
- Gracefully handles servers that don't support resources/prompts
When the Consolidator compresses old session messages into history.jsonl,
those messages are immediately removed from the LLM's context. Dream
processes history.jsonl into long-term memory (memory.md) on a cron
schedule (default every 2h), creating a window where compressed content
is invisible to the LLM.
This change closes the gap by injecting unprocessed history entries
(history.jsonl entries not yet consumed by Dream) directly into the
system prompt as "# Recent History".
Key design notes:
- Uses read_unprocessed_history(since_cursor=last_dream_cursor) so only
entries not yet reflected in long-term memory are included, avoiding
duplication with memory.md
- No overlap with session messages: Consolidator advances
last_consolidated before returning, so archived messages are already
removed from get_history() output
- Token-safe: Consolidator's estimate_session_prompt_tokens calls
build_system_prompt via the same build_messages function, so the
injected entries are included in token budget calculations and will
trigger further consolidation if needed
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
supports_streaming already handles dict configs via isinstance check;
only is_allowed() fails with plain dicts. Narrow the explanation.
Made-with: Cursor
Extends reasoning_effort to accept 'adaptive' in addition to
low/medium/high. When set, uses Anthropic's type: 'adaptive'
thinking API instead of a fixed budget, letting the model decide
when and how much to think per turn.
Also auto-enables interleaved thinking between tool calls on
claude-sonnet-4-6 and claude-opus-4-6.
Usage:
"reasoning_effort": "adaptive" in agents.defaults config
* feat(dream): enhance memory cleanup with staleness detection
- Phase 1: add [FILE-REMOVE] directive and staleness patterns (14-day
threshold, completed tasks, superseded info, resolved tracking)
- Phase 2: add explicit cleanup rules, file paths section, and deletion
guidance to prevent LLM path confusion
- Inject current date and file sizes into Phase 1 context for age-aware
analysis
- Add _dream_debug() helper for observability (dream-debug.log in workspace)
- Log Phase 1 analysis output and Phase 2 tool events for debugging
Tested with glm-5-turbo: MEMORY.md reduced from 149 to 108-129 lines
across two rounds, correctly identifying and removing weather data,
detailed incident info, completed research, and stale discussions.
* refactor(dream): replace _dream_debug file logger with loguru
Remove the custom _dream_debug() helper that wrote to dream-debug.log
and use the existing loguru logger instead. Phase 1 analysis is logged
at debug level, tool events at info level — consistent with the rest
of the codebase and no extra log file to manage.
* fix(dream): make stale scan independent of conversation history
Reframe Phase 1 from a single comparison task to two independent
tasks: history diff AND proactive stale scan. The LLM was skipping
stale content that wasn't referenced in conversation history (e.g.
old triage snapshots). Now explicitly requires scanning memory files
for staleness patterns on every run.
* fix(dream): correct old_text param name and truncate debug log
- Phase 2 prompt: old_string -> old_text to match EditFileTool interface
- Phase 1 debug log: truncate analysis to 500 chars to avoid oversized lines
* refactor(dream): streamline prompts by separating concerns
Phase 1 owns all staleness judgment logic; Phase 2 is pure execution
guidance. Remove duplicated cleanup rules from Phase 2 since Phase 1
already determines what to add/remove. Fix remaining old_string -> old_text.
Total prompt size reduced ~45% (870 -> 480 tokens).
* fix(dream): add FILE-REMOVE execution guidance to Phase 2 prompt
Phase 2 was only processing [FILE] additions and ignoring [FILE-REMOVE]
deletions after the cleanup rules were removed. Add explicit mapping:
[FILE] → add content, [FILE-REMOVE] → delete content.
- Move _tool_hint implementation from loop.py to nanobot/utils/tool_hints.py
- Keep thin delegation in AgentLoop._tool_hint for backward compat
- Update test imports to test format_tool_hints directly
Made-with: Cursor
The lark-oapi client requires token types to be explicitly configured
so that the SDK can obtain and attach the tenant_access_token to raw
requests. Without this, `_fetch_bot_open_id()` would fail with
"Missing access token for authorization" because the token had not
been provisioned at the time of the call.
On Windows, certain Unicode input (emoji, mixed-script text, surrogate
pairs) causes prompt_toolkit's FileHistory to crash with
UnicodeEncodeError when writing the history file.
Fix: wrap FileHistory with a _SafeFileHistory subclass that sanitizes
surrogate characters before writing, replacing invalid sequences instead
of crashing.
Fixes#2846
Add CI step to detect unused imports (F401) and unused variables (F841)
with ruff. Clean up existing violations:
- Remove unused Consolidator import in agent/__init__.py
- Remove unused re import in agent/loop.py
- Remove unused Path import in channels/feishu.py
- Remove unused ContentRepositoryConfigError import in channels/matrix.py
- Remove unused field and CommandHandler imports in channels/telegram.py
- Remove unused exception variable in channels/weixin.py
len(content) counts Unicode code points, not UTF-8 bytes. For non-ASCII
content such as Chinese or emoji, the reported count would be lower than
the actual bytes written to disk, which is misleading to the agent.
The pydantic to_camel function generates 'e2EeEnabled' (treating 'ee'
as a word boundary) for the field 'e2ee_enabled'. Users writing
'e2eeEnabled' in their config get the default value instead.
Fix: add explicit alias='e2eeEnabled' to override the incorrect
auto-generated alias. Both 'e2eeEnabled' and 'e2ee_enabled' now work.
Fixes#2851
The _parse_tavily_usage implementation was updated to use the real
{account: {plan_usage, plan_limit, ...}} structure, but the tests
still used the old flat {used, limit, breakdown} format.
Made-with: Cursor
The Tavily /usage endpoint returns a nested "account" object with
plan_usage/plan_limit/search_usage/etc fields, not the flat structure
with used/limit/breakdown that was assumed. This caused all usage
values to be None.
The /status command tried to access web search config via
`loop.config.tools.web.search`, but AgentLoop has no `config` attribute.
This caused the search usage lookup to silently return None, so web
search provider usage was never displayed.
Fix: use `loop.web_config.search` which is the actual attribute
set during AgentLoop.__init__.
