BlueFish AIAQI AI Intelligence
⌕
Official documentationNot tested by BlueFish

Getting started with local coding agents

Verify the model, then practice permissions and review on reversible code changes.

Last checked:
A quick glossary

Models generate and reason. Agents use tools to work on tasks. A harness manages an agent’s environment. MCP connects tools and data. Skills are reusable task guidance.

Current task

Getting started with local coding agents

Estimated time

Plan 2–4 practice sessions; an editorial estimate, not measured duration.

Limits & boundaries

Wrong port, unloaded model, insufficient resources or incorrect model ID.

Chat works but tools are unreliable, or context is insufficient.

Preview configuration changes, weak tool capability or web search requesting cloud authentication.

Validation

Change one heading in a test repo, review and revert with Git, explaining every command approval.

Next steps

Complete the linked recipes’ verification items before adding one capability; repeat source checks when dependencies change.

Cost

Not verified

Price not verified; check software, model, API, hardware and hosting costs separately.

Implementation steps

  1. 01
    Send a short official API request, record model ID, errors and latency, and avoid private files.
  2. 02
    Complete the local chat recipe and prepare a test Git repo without secrets.
  3. 03
    Select Ollama and the actual model in Cline without broadening file or command permissions.
  4. 04
    Ask it to explain a file, then change one sentence and review the diff.
  5. 05
    Manually run appropriate checks and record success and failure before expanding the task.

Implementation steps

  1. Send a short official API request, record model ID, errors and latency, and avoid private files.

  2. Complete the local chat recipe and prepare a test Git repo without secrets.

  3. Select Ollama and the actual model in Cline without broadening file or command permissions.

  4. Ask it to explain a file, then change one sentence and review the diff.

  5. Manually run appropriate checks and record success and failure before expanding the task.

Original sources

  1. Ollama official documentation ↗

    docs.ollama.com · 2026-10-15 · Official documentation

  2. LM Studio official documentation ↗

    lmstudio.ai · 2026-10-15 · Official documentation

  3. Cline official documentation ↗

    github.com · 2026-10-15 · Official documentation

  4. DeepSeek Harness official documentation ↗

    github.com · 2026-10-15 · Official documentation

  5. Ollama with Harness ↗

    github.com · 2026-10-15 · Official documentation

  6. Harness model configuration ↗

    github.com · 2026-10-15 · Official documentation

Connected resources

HarnessFree software; check license and compute costs

DeepSeek Harness

An official open-source agent harness from DeepSeek AI with a plugin architecture, in developer preview.

Official documentation
AgentFree software; check license and compute costs

Cline

A coding agent supporting multiple providers, Ollama, LM Studio and MCP.

Official documentation
Local runtimeFree software; check license and compute costs

Ollama

A local model runtime and API tool supporting a subset of OpenAI endpoints.

Official documentation
Local runtimeFree software; check license and compute costs

LM Studio

A desktop local-model tool with compatible Chat Completions and Responses endpoints.

Official documentation