ollama_research

Research over files. FLAGSHIP. Answer a question grounded in specific files. Takes FILE PATHS (not raw text) — reads and chunks locally, returns a digest with validated citations. Use this to understand a repo/doc without burning Claude context on the full content. Citations outside source_paths are stripped server-side; cited line_ranges that point past EOF have the range dropped (path is kept) with a warning. Returns honest grounding signals: weak: true when an answer has zero validated citations (likely ungrounded); abstained: true when the model explicitly refused (citations cleared); sources_address_question is tri-state (null when unknown).

Parameters

Parameter	Type	Required	Default	Description
question	string	yes	—	The question to answer.
source_paths	any	yes	—	Files the answer must be grounded in. Nothing outside this list is allowed as a source.
max_words	integer	no	—	Target answer length in words (default 300).
per_file_max_chars	integer	no	—	Chars to read per file (default 40k).
model	string	no	—	Optional per-call model override. When provided, overrides the tool’s tier-resolved model for this call. The tier’s timeout (TIER_TIMEOUT_MS) still applies. On timeout, fallback uses the tier-resolved model, NOT the override. Use for receipt-backed orchestration that requires explicit model identity (e.g., research-os reviewer profiles).

Full input schema

The JSON Schema below is generated from the same zod schema the server validates against at the wire — it cannot drift from runtime behavior.

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "question": {
      "type": "string",
      "minLength": 1,
      "description": "The question to answer."
    },
    "source_paths": {
      "description": "Files the answer must be grounded in. Nothing outside this list is allowed as a source."
    },
    "max_words": {
      "description": "Target answer length in words (default 300).",
      "type": "integer",
      "minimum": 20,
      "maximum": 1500
    },
    "per_file_max_chars": {
      "description": "Chars to read per file (default 40k).",
      "type": "integer",
      "minimum": 1000,
      "maximum": 200000
    },
    "model": {
      "description": "Optional per-call model override. When provided, overrides the tool's tier-resolved model for this call. The tier's timeout (TIER_TIMEOUT_MS) still applies. On timeout, fallback uses the tier-resolved model, NOT the override. Use for receipt-backed orchestration that requires explicit model identity (e.g., research-os reviewer profiles).",
      "type": "string",
      "minLength": 1
    }
  },
  "required": [
    "question",
    "source_paths"
  ]
}

Envelope

Every tool returns the standard envelope — tier_used, model, tokens_in/tokens_out, elapsed_ms, backend provenance, warnings. See Envelope & tiers.

Source

Schema + handler: src/tools/research.ts · registration: src/index.ts

---

Back to the Tool Reference.