Add Ollama Provider
Routes a NanoClaw agent group to a local Ollama model via env var overrides.
Test report
- Verdict
- Tested · Didn't pass
- Tested
- Jul 11, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 740818b
SKILL.md's core steps don't match the codebase its own documented install clones: step 1a/1b edit a function `readContainerConfig` and insert after a `NANOCLAW_MCP_SERVERS` anchor that exist nowhere in the repo (verified on both main and providers branches), and step 3's hand-edited groups/<folder>/container.json is silently regenerated from the container_configs DB table by materializeContainerJson() on every container spawn — so the change never persists, and the DB row has no env/blockedHosts columns. The description's 'no provider code changes are needed' also contradicts the skill's own steps 1a/1b, which add TypeScript fields and docker-arg wiring (required on a fresh clone). On the adapted checkable task the A/B was a wash: skill and base arms produced byte-identical ContainerConfig edits and container.json/settings.json, because the task is well-specified and the skill's literal anchors had to be abandoned — the skill added no lift and, followed verbatim, would have misdirected. Full end-to-end verify (build/restart, live Ollama, Docker, provisioned agent-group DB row) is blocked in a fresh workdir per the skill's own prerequisites.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 4/5
- Output vs. baseline 4/10
- Docs & honesty 1/5
What Add Ollama Provider does
Route a NanoClaw agent group to a local Ollama model instead of the Anthropic API. Ollama speaks the Anthropic API natively (v1/messages), so no provider code changes are needed — just env var overrides and a model setting. Use when the user wants to run their agent locally, cut API costs, or experiment with open-weight models. See docs/ollama.md for background.
How to install Add Ollama Provider
git clone https://github.com/nanocoai/nanoclaw
cd nanoclaw
mkdir -p ~/.claude/skills
cp -r .claude/skills/add-ollama-provider ~/.claude/skills/add-ollama-provider
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Add Ollama Provider
-
/add-ollama-providerRoutes a NanoClaw agent group to a local Ollama model via env var overrides.
It also activates on plain-language prompts like these:
-
Switch this agent to run on a local Ollama model -
Configure NanoClaw to use Ollama instead of Anthropic -
Point the agent group at a local open-weight model
Frequently asked questions
- Is the Add Ollama Provider skill free?
- Yes. The skill itself is free from nanocoai/nanoclaw. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Add Ollama Provider work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 11, 2026. Verdict: Tested · Didn't pass. SKILL.md's core steps don't match the codebase its own documented install clones: step 1a/1b edit a function `readContainerConfig` and insert after a `NANOCLAW_MCP_SERVERS` anchor that exist nowhere in the repo (verified on both main and providers branches), and step 3's hand-edited groups/<folder>/container.json is silently regenerated from the container_configs DB table by materializeContainerJson() on every container spawn — so the change never persists, and the DB row has no env/blockedHosts columns. The description's 'no provider code changes are needed' also contradicts the skill's own steps 1a/1b, which add TypeScript fields and docker-arg wiring (required on a fresh clone). On the adapted checkable task the A/B was a wash: skill and base arms produced byte-identical ContainerConfig edits and container.json/settings.json, because the task is well-specified and the skill's literal anchors had to be abandoned — the skill added no lift and, followed verbatim, would have misdirected. Full end-to-end verify (build/restart, live Ollama, Docker, provisioned agent-group DB row) is blocked in a fresh workdir per the skill's own prerequisites.
- How do I install Add Ollama Provider?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Add Ollama Provider with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.