Implements KnowledgeSearchTool registered under "knowledge_search" that wraps
KnowledgeStore with optional filters (source, doc_type, author, since, until,
top_k) and formats results with source attribution for agent consumption.
Includes 8 unit tests covering basic search, filter-by-source, filter-by-author,
no-results, empty-query, no-store, spec parameters, and registry checks.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Implements GmailConnector registered under 'gmail' in ConnectorRegistry,
a shared oauth.py helper (build_google_auth_url, load/save/delete_tokens)
reusable by Drive/Calendar/Contacts, and 7 fully mocked pytest tests
covering auth state, sync document extraction, disconnect, mcp_tools,
and registry lookup.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Implements ObsidianConnector (filesystem auth, no OAuth) that walks a vault
directory for .md/.markdown/.txt files, parses YAML frontmatter via a
dependency-free parser, skips hidden dirs and binary files, and yields
Document objects with doc_type="note" and obsidian:// deep-link URLs.
Exposes an obsidian_search_notes MCP ToolSpec. 9 tests cover connection
state, vault traversal, hidden-dir/binary filtering, frontmatter extraction,
and registry wiring.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Introduces SyncEngine that wraps IngestionPipeline with a SQLite state
database (sync_state.db) for checkpoint/resume: cursors and item counts
are persisted after every 100-document batch and on completion/error.
Adds 4 tests covering single-connector ingestion, checkpoint accuracy,
unsynced-connector None return, and multi-connector source filtering.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Implements IngestionPipeline that deduplicates Documents by doc_id (both
in-memory and by loading existing doc_ids from the store on init), chunks
content via SemanticChunker, and persists all chunks with full provenance
metadata to KnowledgeStore. 8 tests cover single-doc ingestion, dedup across
calls and batches, persistence across pipeline instances, long-doc multi-chunk
splitting, atomic event chunking, multi-source filtering, and return-value accuracy.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Implements SemanticChunker that splits text based on doc_type (event/contact
as single chunks, email on reply boundaries, message on double-newlines,
document/note on ## headings → paragraphs → sentences). ChunkResult carries
sequential 0-based indexes and inherits parent metadata; section headings are
added as chunk metadata. 16 tests covering all splitting strategies.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Implements KnowledgeStore (extends MemoryBackend) for Deep Research with a
rich SQLite/FTS5 schema supporting source, doc_type, author, participants,
timestamp, thread_id, url, and chunk_index columns; BM25 ranking via FTS5
with porter unicode61 tokenizer; filtered retrieval by source, doc_type,
author, since, and until; MEMORY_STORE/MEMORY_RETRIEVE event emission; WAL
journal mode; and 15 isolated tests using tmp_path fixtures.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Document that pygemma v0.1.3 completion() does not accept
temperature/max_tokens (params accepted for ABC compliance)
- Add model-mismatch warning to stream() for consistency with generate()
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>
AgenticRunner dispatches _run_body() to a ThreadPoolExecutor when a
task environment is present (for Playwright compatibility). The
TraceStore connection was created on the main thread, causing
"SQLite objects created in a thread can only be used in that same
thread" on the first agentic eval query.
Pass check_same_thread=False to sqlite3.connect(), consistent with
SchedulerStore, AgentManager, SessionStore, and TelemetryStore which
already use this flag.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Store workspace_path in record.metadata from AgenticRunner so task
envs can reuse the agent's workspace instead of creating a separate
temp dir the agent can't see
- PinchBenchTaskEnv now uses the existing workspace when available,
copies fixtures there, and sets CWD so file_read/file_write resolve
relative paths correctly
- Wrap grading in try/except so LLM judge failures don't crash the
entire run (graceful degradation to score 0.0)
- Restore CWD on exit and only clean up self-owned workspaces
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds a truncated (10KB max) string representation of the tool result's
content to the TOOL_CALL_END event payload so grading functions (e.g.
PinchBench) can inspect what each tool returned.
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>
- Merge Quick Start, Docker, and Development into single Quick Start
- Point to docs site for full documentation (Docker, cloud, dev setup)
- Add Contributing section with dev setup and roadmap pointer
Co-Authored-By: Claude Opus 4.6 (1M context) <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>
These are Claude-generated working documents that should not
be checked into the repository.
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>
Thread prompt_tokens_evaluated through the telemetry stack so
savings calculations use the right token count for each metric:
- Dollar cost: full prompt_tokens (what cloud providers charge)
- FLOPs/energy: prompt_tokens_evaluated (actual compute with KV
cache — subsequent turns only re-evaluate new tokens)
Ollama reports both: prompt_eval_count (cache-aware) and we estimate
full count from messages. OpenAI-compat engines report full count
only (KV caching is transparent in their API).
Changes: TelemetryRecord, store schema, aggregator, engines,
savings calculation, /v1/savings route.
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>
