archeious/marchwarden
1
0
Fork 0
Code Issues 14 Pull requests Projects Releases Packages Wiki Activity Actions
16 commits 22 branches 0 tags 442 KiB
Commit graph

7 commits

Author SHA1 Message Date
Jeff Smith
5d894d9e10 M1.4: MCP server wrapping web researcher 
FastMCP server exposing a single 'research' tool:
- Delegates to WebResearcher with keys from ~/secrets
- Accepts question, context, depth, max_iterations, token_budget
- Returns full ResearchResult as JSON
- Configurable model via MARCHWARDEN_MODEL env var
- Runnable as: python -m researchers.web

4 tests: secret reading, JSON response validation, default parameters.

Refs: archeious/marchwarden#1

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
 2026-04-08 14:41:13 -06:00
Jeff Smith
ae9c11a79b Add OpenQuestion to research contract 
New field on ResearchResult: open_questions — follow-up questions that
emerged from the research itself. Distinct from gaps (backward: what
failed) and discovery_events (sideways: what's lateral). Open questions
look forward: 'based on what I found, this needs deeper investigation.'

- OpenQuestion model: question, context, priority (high/medium/low),
  source_locator
- Updated agent synthesis prompt to produce open_questions
- Updated agent result builder to parse open_questions from JSON
- 3 new tests for OpenQuestion model
- Updated existing tests for new field

77 tests passing.

Refs: archeious/marchwarden#1

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
 2026-04-08 14:37:30 -06:00
Jeff Smith
7cb3fde90e M1.3: Inner agent loop with tests 
WebResearcher — the core agentic research loop:
- Tool-use loop: Claude decides when to search (Tavily) and fetch (httpx)
- Budget enforcement: stops at max_iterations or token_budget
- Synthesis step: separate LLM call produces structured ResearchResult JSON
- Fallback: valid ResearchResult even when synthesis JSON is unparseable
- Full trace logging at every step (start, search, fetch, synthesis, complete)
- Populates all contract fields: raw_excerpt, categorized gaps,
  discovery_events, confidence_factors, cost_metadata with model_id

9 tests: complete research loop, budget exhaustion, synthesis failure
fallback, trace file creation, fetch_url tool integration, search
result formatting.

Refs: archeious/marchwarden#1

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
 2026-04-08 14:29:27 -06:00
Jeff Smith
cef08c8984 M1.2: Trace logger with tests 
TraceLogger produces JSONL audit logs per research() call:
- One file per trace_id at ~/.marchwarden/traces/{trace_id}.jsonl
- Each line is a self-contained JSON object (step, action, timestamp, decision)
- Supports arbitrary kwargs (url, content_hash, query, etc.)
- Lazy file handle, flush after each write, context manager support
- read_entries() for replay and testing

15 tests: file creation, step counting, JSONL validity, kwargs,
timestamps, flush behavior, multiple independent traces.

Refs: archeious/marchwarden#1

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
 2026-04-08 14:21:10 -06:00
Jeff Smith
a5bc93e275 M1.1: Search and fetch tools with tests 
- tavily_search(): Tavily API wrapper returning SearchResult dataclasses
  with content hashing (raw_content preferred, falls back to summary)
- fetch_url(): async URL fetch with HTML text extraction, content hashing,
  and graceful error handling (timeout, HTTP errors, connection errors)
- _extract_text(): simple HTML → clean text (strip scripts/styles/tags,
  decode entities, collapse whitespace)
- _sha256(): SHA-256 content hashing with 'sha256:' prefix for traces

18 tests: hashing, HTML extraction, mocked Tavily search, mocked async
fetch (success, timeout, HTTP error, hash consistency).

Refs: archeious/marchwarden#1

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
 2026-04-08 14:17:18 -06:00
Jeff Smith
1b0f86399a M0.3: Implement contract v1 Pydantic models with tests 
All Research Contract types as Pydantic models:
- ResearchConstraints (input)
- Citation with raw_excerpt (output)
- GapCategory enum (5 categories)
- Gap with structured category (output)
- DiscoveryEvent (lateral findings)
- ConfidenceFactors (auditable scoring inputs)
- CostMetadata with model_id (resource tracking)
- ResearchResult (top-level contract)

32 tests: validation, bounds checking, serialization roundtrips,
JSON structure verification against contract spec.

Refs: archeious/marchwarden#1

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
 2026-04-08 14:00:45 -06:00
Jeff Smith
deb124ed29 Initial project structure and scaffolding 
- Directory layout: researchers/web/, orchestrator/, cli/, docs/wiki/
- README with quick start and vision
- CONTRIBUTING with workflow and testing guidelines
- pyproject.toml with dependencies and build config
- .gitignore for Python projects

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
 2026-04-08 11:57:15 -06:00