Skip to content

ollama_repo_pack

Repo pack (writes artifact). PACK. Runs the full repo ONBOARDING job: corpus_search (if corpus given) → repo_brief → targeted ollama_extract (narrow onboarding schema: packages, entrypoints, scripts, config_files, exposed_surfaces, runtime_hints) → deterministic markdown+JSON artifact on disk. Corpus-first posture: when a corpus is given, it’s the main working surface alongside source_paths. Not repo Q&A — this is get me onboarded fast with a stable operator artifact. Response is compact; full brief + extracted facts live in the artifact.

Parameter Type Required Default Description
source_paths string[] yes Files the onboarding brief should read (README, key src entries, package/manifest, docs). Required — these are the repo’s concrete source-of-truth.
corpus string no Optional named corpus (e.g. ‘handbook’, ‘doctrine’) for cross-cutting architecture context. When given, queried as the pack’s main working surface alongside source_paths.
corpus_query string no Corpus query (defaults to ‘repo architecture and surfaces’). (max 200 chars — deliberately shorter than ollama_corpus_search.query’s 1000, because this is a short retrieval prompt, not a document.)
title string no Short human title — used in the artifact header and filename slug. Defaults to the repo thesis head.
artifact_dir string no Directory to write the repo.md + repo.json artifact pair. Defaults to ~/.ollama-intern/artifacts/repo/.
allowed_roots string[] no Absolute directories artifact_dir may live under when it is not inside INTERN_ARTIFACT_DIR. Same dual-declaration as ollama_artifact_export_to_path.
confirm_write boolean no Required when the artifact pair would land on a protected path (.git/, SECURITY.md, memory/, …). Same gate as ollama_draft.
per_file_max_chars integer no Chars per source file (default 20k).
max_key_surfaces integer no Cap on key_surfaces (default 8).
max_risk_areas integer no Cap on risk_areas (default 5).
max_read_next integer no Cap on read_next (default 8).
backend "cloud" \ "local" no

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

{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"type": "object",
"properties": {
"source_paths": {
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"description": "Files the onboarding brief should read (README, key src entries, package/manifest, docs). Required — these are the repo's concrete source-of-truth."
},
"corpus": {
"description": "Optional named corpus (e.g. 'handbook', 'doctrine') for cross-cutting architecture context. When given, queried as the pack's main working surface alongside source_paths.",
"type": "string",
"pattern": "^[a-zA-Z0-9_-]+$"
},
"corpus_query": {
"description": "Corpus query (defaults to 'repo architecture and surfaces'). (max 200 chars — deliberately shorter than ollama_corpus_search.query's 1000, because this is a short retrieval prompt, not a document.)",
"type": "string",
"minLength": 1,
"maxLength": 200
},
"title": {
"description": "Short human title — used in the artifact header and filename slug. Defaults to the repo thesis head.",
"type": "string",
"minLength": 1,
"maxLength": 120
},
"artifact_dir": {
"description": "Directory to write the repo.md + repo.json artifact pair. Defaults to ~/.ollama-intern/artifacts/repo/.",
"type": "string",
"minLength": 1
},
"allowed_roots": {
"description": "Absolute directories artifact_dir may live under when it is not inside INTERN_ARTIFACT_DIR. Same dual-declaration as ollama_artifact_export_to_path.",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"confirm_write": {
"description": "Required when the artifact pair would land on a protected path (.git/, SECURITY.md, memory/, ...). Same gate as ollama_draft.",
"type": "boolean"
},
"per_file_max_chars": {
"description": "Chars per source file (default 20k).",
"type": "integer",
"minimum": 1000,
"maximum": 200000
},
"max_key_surfaces": {
"description": "Cap on key_surfaces (default 8).",
"type": "integer",
"minimum": 1,
"maximum": 20
},
"max_risk_areas": {
"description": "Cap on risk_areas (default 5).",
"type": "integer",
"minimum": 1,
"maximum": 10
},
"max_read_next": {
"description": "Cap on read_next (default 8).",
"type": "integer",
"minimum": 1,
"maximum": 15
},
"backend": {
"description": "Optional per-call backend directive (v2.9). 'cloud' escalates THIS call to Ollama Cloud — works in cloud standby (OLLAMA_API_KEY set, OLLAMA_CLOUD_PRIMARY unset) and cloud-primary modes; errors with CLOUD_NOT_CONFIGURED when no cloud is configured (never silently runs local while claiming escalation). 'local' pins the call to the local backend (zero egress) even under cloud-primary. Omit for the mode default. Escalated calls disclose egress loudly and carry backend provenance on the envelope. Synthesis quality on this tool scales sharply with model class — on a repo-scale job the local 8B is often the binding constraint, not the evidence. Tools where a local model is genuinely adequate (ollama_summarize_fast, ollama_classify, ollama_triage_logs, and every no-LLM tool) deliberately do NOT offer this field. SCOPE: this directive reaches the brief synthesis step (ollama_repo_brief, Deep tier) ONLY — evidence assembly, the targeted extract (Workhorse structured fill) and the artifact write run on the mode default regardless. A pack is a mixed-cost pipeline; escalating all of it would bill a flagship for work a local model does fine.",
"type": "string",
"enum": [
"cloud",
"local"
]
}
},
"required": [
"source_paths"
]
}

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

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


Back to the Tool Reference.