I tested the same strict document-review prompt directly in Ollama using `ollama run gemma4:e4b`, and the model followed the required structure much better. Direct Ollama output used the requested headings: ```text DRAFT — HUMAN REVIEW REQUIRED 1. Executive Summary 2. Clarity Issues 3. Missing Information 4. Risky or Overstated Claims 5. Operational Gaps 6. Suggested Improvements 7. Recommended Next Action 8. Final Review Status ``` But when I run a similar workflow through OpenClaw, the agent keeps drifting into freeform summaries and adds its own headings, even after I tell it not to. So now it looks like the model can follow the template directly, but OpenClaw may be adding session/profile/runtime behavior that changes the output style. Has anyone seen this with OpenClaw local agents? I’m trying to figure out: 1. How to fully reset/clear OpenClaw session context 2. Whether the default agent profile is adding hidden behavior 3. Whether there is a better no-tools/local-only config 4. Whether I should create a separate strict “document review” agent instead of using the main agent 5. Whether OpenClaw has a setting for system prompts or fixed output templates Current stack: * Windows desktop * Ollama working * Gemma direct test working * OpenClaw gateway reachable * Web-search plugin disabled * Local-only mode working * Issue is strict template adherence inside OpenClaw, not direct Ollama

Hey everyone — I’m testing a local OpenClaw + Ollama + Gemma setup on Windows and wanted to see if anyone else has run into this behavior. **Environment:** * OS: Windows desktop * Ollama: installed and working * Model: `gemma4` / `gemma4:e4b` * OpenClaw: installed and gateway reachable * Local model responds correctly when running directly through Ollama **What works:** * `ollama --version` works * `ollama list` shows installed models * `ollama run gemma4:e4b` responds correctly * `openclaw status --deep` shows the gateway reachable and event loop OK * OpenClaw opens and the local agent can respond after disabling the web-search plugin issue **Original issue:** When using OpenClaw, I was getting this error: ```text No callable tools remain after resolving explicit tool allowlist tools.allow: *, ollama_web_search, ollama_web_fetch no registered tools matched Fix the allowlist or enable the plugin that registers the requested tool. ``` It looked connected to the `openclaw-web-search` plugin. My config had: ```json "plugins": { "allow": [ "openclaw-web-search", "ollama", "memory-core" ], "entries": { "ollama": { "enabled": true }, "openclaw-web-search": { "enabled": true } } }, "tools": { "alsoAllow": [ "ollama_web_search", "ollama_web_fetch" ], "web": { "fetch": { "enabled": false }, "search": { "enabled": false } } } ``` I temporarily fixed the tool error by disabling `openclaw-web-search` and removing `ollama_web_search` / `ollama_web_fetch` from the allowlist so I could run in local-only mode. **Current issue:** Now the local agent responds, but it has trouble consistently following a strict fixed output template. For example, I gave it a required review-report format with exact headings, but it keeps adding its own headings or summarizing freely instead of following the requested structure. I’m trying to build a local-only, draft-only business document review workflow where the agent must: * Review only the text provided * Not invent context * Not browse the web

Let's get to know each other! Comment below sharing where you are in the world, a career goal you have, and something you like to do for fun. 😊

Anthropic just launched Managed Agents, and it's basically a way to build and deploy AI agents directly in their cloud without any infrastructure setup. In this video I walk through how they work, what they cost, and what I actually built with them. I also break down why I'm a little disappointed and when you should (or shouldn't) use them compared to tools like Claude Code, trigger.dev, or OpenClaw.

1) 🚫 No Business Promotions → NO “DM me for…” or "Comment 'Automation'" posts. 2) 🔗 No Linking Your Own Community/YouTube Videos 3) 🏷️ Title Specifically 4) 🔍 Search for Help First (searchbar) 5) 🙌 Stay Respectful 6) ❌ Enforced Clean‑Up Posts that break these rules will be removed without warning. If you ever have questions, feel free to ask. Let’s make this the best AI Automation community out there by sharing, collaborating, and learning together. 🚀