_apply_toml_section only normalized TOML arrays to comma-separated strings
for real dataclass fields, but backward-compat property setters like
reward_weights also expect string input. When a user's config.toml had
an array value for a property-backed attribute, the raw list was passed
to the setter which called .split(",") on it, causing:
'list' object has no attribute 'split'
This also hardens serve.py against the same issue when reading
config.agent.tools.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Merge main into fix/ssrf-check, keeping both the auto-recover
logic for error-state agents and the async streaming support
for the send_message endpoint.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(evals): PinchBench harness fixes — scores from 26% to 84%
Multiple infrastructure bugs prevented PinchBench from producing
accurate scores. This commit fixes the eval harness so model scores
match the official leaderboard (Qwen3.5-397B: 26% → 84%, GPT-5.4:
5% → 58%).
Key fixes:
- EvalRunner: wrap generation in PinchBenchTaskEnv context so workspace
files persist through grading (was deleting before scorer ran)
- EvalRunner: detect task_env datasets and force sequential processing
(CWD changes aren't thread-safe)
- EvalRunner: fix episode_mode auto-detection to check for real
iter_episodes() override instead of hasattr() (always True)
- Scorer: use "params" field in transcripts to match PinchBench grade()
functions (was "arguments")
- Scorer: add None guard in _trace_to_transcript for tool_calls
- Scorer: capture final assistant text response in transcript so
text-only tasks (like sanity check) can be graded
- Scorer: add _tool_results_to_transcript() helper for EvalRunner path
- native_openhands: add native function-calling support (tools=
parameter) with text-based fallback, matching monitor_operative
- Tools: add Python fallbacks for file_read, file_write, shell_exec,
calculator, think, http_request when openjarvis_rust unavailable
- Security: add Python fallback for is_sensitive_file()
- Config: add "pinchbench" to KNOWN_BENCHMARKS
- Add PinchBench eval configs for Qwen3.5-397B and GPT-5.4
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(evals): fix hybrid grading crash when grading_weights is None
The 4 tasks with grading_type=hybrid (task_10, task_13, task_16_market,
task_22) crashed because grading_weights was explicitly None in task
metadata. dict.get("key", default) returns None (not the default) when
the key exists with value None. Use `or` to coalesce None to defaults.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(evals): handle MagicMock datasets in runner type checks
The iter_episodes and create_task_env type identity checks fail with
AttributeError when the dataset is a MagicMock (used in tracker tests).
Wrap in try/except to default to False for non-DatasetProvider objects.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Move mid-file imports to top of test file to resolve E402 violations and apply ruff formatting to both gemma_cpp.py and test_gemma_cpp.py.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds a `tool_calls` list field to `TurnTrace` capturing name, arguments,
and result for each tool invocation, enabling PinchBench transcript integration.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The test_init_creates_config test in test_cli.py was not updated
when the download prompt and privacy hook were added to jarvis init.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Wraps long CliRunner.invoke() calls, removes unused pytest imports,
fixes import ordering, and removes unused variables in test_scan.py.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add slots=True to ScanResult dataclass
- Fix extra space in IPv6 port f-string
- Add missing tests: check_remote_access, check_icloud_sync, run_quick
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Registers the new scan command in the CLI. Adds a lightweight
privacy check at the end of jarvis init that runs disk encryption
and cloud sync checks, with pointer to jarvis scan for full audit.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Prompts user to download recommended model during jarvis init.
Adds --no-download flag for CI. Shows helpful message when no
model fits available memory. Adds exo/nexa next-steps text.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The FLOPs formula changed from quadratic (P*N*(N+1)) to linear
(2*P*T_evaluated) with KV-cache awareness. Update the test
expectation from ~100x to ~10x for 10x token increase.
Refactors model pull into reusable ollama_pull() function. Adds
--engine flag to support llamacpp (GGUF) and mlx (HuggingFace)
downloads via huggingface-cli, with FileNotFoundError handling.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
New PrivacyScanner class with checks for disk encryption (FileVault/LUKS),
MDM profiles, cloud sync agents, network exposure, and screen recording.
Supports macOS and Linux with graceful skip on missing tools.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Fixes recommend_model() returning empty string on Apple Silicon
when MLX is the recommended engine. Also adds gguf_file and
mlx_repo download metadata, and estimated_download_gb helper.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The test_init_defaults test expected OperativeAgent class defaults
(temperature=0.3) but didn't mock load_config(), so when config loaded
successfully it returned the global default (0.7) instead.
Fix: mock load_config to raise, so the test properly validates the
class-level _default_temperature/max_tokens/max_turns fallback path.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Eight tool modules (file_write, apply_patch, git_tool, db_query,
pdf_tool, image_tool, audio_tool, knowledge_tools) were missing from
openjarvis/tools/__init__.py. Their @ToolRegistry.register() decorators
never fired, so the /v1/tools endpoint never returned them and the
web UI agent wizard showed an incomplete tool list.
Add the missing imports and a regression test that checks all expected
tool names are in the registry after package import.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat: add query complexity analyzer with CLI and UI integration
Classify incoming queries by difficulty (trivial→very_complex) to
suggest appropriate token budgets for local vs. cloud routing.
- Add score_complexity() with weighted signals (length, code, math,
reasoning, multi-step, creative) and token tier mapping
- Wire into `jarvis ask`: auto-suggest max_tokens when not set by user,
show complexity in --profile output, log at DEBUG level
- Add complexity metadata to /v1/chat/completions API response
- Display complexity tier and score in frontend XRayFooter
- Extend RoutingContext with complexity_score, suggested_max_tokens,
has_reasoning fields
- Update HeuristicRouter to route on complexity_score instead of
raw query_length
- Remove duplicated regex patterns from router.py (use complexity
module as single source of truth)
- Add 30 unit tests for complexity module
- Fix existing router tests for new complexity-based routing
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: pass temperature and max_tokens from UI settings to backend
The settings page stores temperature and max_tokens in the frontend
store, but these values were never included in the chat API request.
The backend Pydantic model defaults max_tokens to 1024 when the field
is absent, which is too low for thinking models like qwen3.5 — they
consume all tokens on reasoning and return empty content.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: pass UI settings to backend and auto-bump max_tokens from complexity
- Pass temperature and max_tokens from frontend Settings store to the
backend API (cherry-picked from fix/ui-max-tokens-passthrough)
- Server-side: bump max_tokens when the complexity analyzer suggests a
higher budget (e.g. for thinking models on complex queries), never
reduce below the client-requested value
- Fixes empty responses with thinking models (e.g. qwen3.5) that
consumed all tokens on internal reasoning
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: raise token budget tiers to prevent empty responses on thinking models
Double all tier budgets (trivial: 512→1024, simple: 1024→2048, etc.)
so that thinking models like qwen3.5 have enough headroom for internal
chain-of-thought plus visible output, even on simple queries.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: CI lint and test failures
- Break long lines in routes.py to satisfy 88-char limit (E501)
- Add intelligence.max_tokens and temperature to mocked config in
test_ask_router.py so complexity analyzer can compare against int
instead of MagicMock
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: line too long in test_complexity.py (E501)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add `stream: bool` parameter to `POST /v1/managed-agents/{id}/messages`.
When `stream=true`, the agent processes the message synchronously and
returns an SSE stream (OpenAI-compatible format) with token-by-token
response, tool result events, and usage metadata.
This enables real-time voice assistants and chat UIs to receive agent
responses as they are generated, rather than polling for completion.
- Extend SendMessageRequest with `stream` field (default: false)
- Add _stream_managed_agent() helper using asyncio.to_thread()
- Build AgentContext from conversation history for multi-turn support
- Emit tool_results as named SSE events
- Persist agent response in DB after streaming completes
- Add 6 new tests covering streaming behavior
- Update agents.md documentation with streaming examples
* feat: add CLI commands: config, registry, tool
* fix: address review feedback on CLI config/registry/tool commands
- Replace non-existent `create_config_template` with `generate_default_toml`
to fix ImportError when config file is missing
- Deduplicate registry maps into shared `_load_registry_map()` helper
- Remove dead `_get_registry_class()` function and its tests
- Route JSON output to stdout for pipeability (`jarvis config show json | jq`)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Jon Saad-Falcon <41205309+jonsaadfalcon@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The energy_wh_saved and flops_saved values were orders of magnitude too
high because a scaling factor that grows linearly with N was applied to
the energy calculation, making it scale as O(N³) instead of O(N²).
Replace the buggy scale-factor approach with a direct FLOP-to-energy
conversion using each provider's per-token constants. Add regression
tests to prevent recurrence.
Closes#95
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat: add MiniMax as cloud inference provider
Add MiniMax M2.5 and M2.5-highspeed as a 5th cloud provider alongside
OpenAI, Anthropic, Google, and OpenRouter. Uses the OpenAI-compatible
API at api.minimax.io/v1 via the existing openai SDK dependency.
Changes:
- Add MiniMax client init, generate, and streaming in CloudEngine
- Add MiniMax models to model catalog with correct pricing
- Add MINIMAX_API_KEY environment variable support
- Add temperature clamping (0.01-1.0) per MiniMax API constraints
- Add 19 unit tests and 3 integration tests
- Update docs and README with MiniMax provider info
* feat: upgrade MiniMax default model to M2.7
- Add MiniMax-M2.7 and MiniMax-M2.7-highspeed to model list
- Set MiniMax-M2.7 as default model (first in list)
- Keep all previous models (M2.5, M2.5-highspeed) as alternatives
- Update pricing table with M2.7 entries
- Update model catalog with M2.7 specs
- Update docs to list all available MiniMax models
- Update unit tests (23 passing) and integration tests (3 passing)
---------
Co-authored-by: Octopus <octo-patch@users.noreply.github.com>
- Catch any Exception from Tavily (not just specific error types)
- Fall back to DuckDuckGo for any error, making the tool more robust
- Fix test mocks to use builtins.__import__ for proper local import mocking
- Simplify test_execute_tavily_error to test generic exception handling
When Tavily API is unavailable (no API key, import error, or API error),
the web_search tool now falls back to DuckDuckGo search instead of
failing. This ensures the tool always works for users without a
Tavily API key.
- Add DuckDuckGo search as fallback using ddgs package
- Catch specific Tavily exceptions (MissingAPIKeyError, InvalidAPIKeyError,
ForbiddenError, UsageLimitExceededError, TimeoutError, BadRequestError)
- Add logger.debug calls to log when falling back to DuckDuckGo
- Use ddgs instead of deprecated duckduckgo-search package name
- Add test for DuckDuckGo fallback result formatting
- Simplify test mocking to use consistent monkeypatch patterns
Closes#81
- Acquire tick BEFORE spawning thread to prevent race condition on concurrent Run Now clicks
- Auto-recover agents in error/needs_attention state when Run Now is clicked
- Auto-recover error-state agents when receiving immediate messages
- End tick on system build failure to avoid stuck running state
- Add test verifying concurrent start_tick raises ValueError
