Preview: Run Coding Agents for a Team: Shared Rules, a Test Set, Nightly Jobs and No Lock-In
Level L05
T20
20 min read
Turn single-person coding-agent set-ups into a team routine: one AGENTS.md every harness reads, a test set that defines done, a nightly CI job that fails loudly, and harness, model and gateway as separate settings.
Matching textT20-L05 · AI harnesses · Level 5 Operator · 20-minute read, plus 2½ hours of doing Run Coding Agents for a Team: Shared Rules, a Test Set, Nightly Jobs and No Lock-In Five people, three harnesses,…
Preview: Set Up a Harness So a Mistake Stays Small: Container, Rules, Keys and Network
Level L04
T20
20 min read
Keep an unattended coding agent small with four layers: a container that sees only the project, a network that reaches only the model gateway, a narrow expiring key, and harness rules for honest mistakes. Includes a five-action safety probe.
Matching textT20-L04 · AI harnesses · Level 4 Integrator · 20-minute read, plus 2 hours of doing Set Up a Harness So a Mistake Stays Small: Container, Rules, Keys and Network The agent runs at night Your…
Preview: Run a Harness on a Model You Control: Local Server or Team Gateway
Level L03
T20
18 min read
Connect OpenCode and DeepSeek Harness to a local model server or a team gateway, set the four values that must match, and test whether a model can act as an agent before colleagues rely on it.
Matching textT20-L03 · AI harnesses · Level 3 Builder · 18-minute read, plus 90 minutes of doing Run a Harness on a Model You Control: Local Server or Team Gateway The data may not leave the building Your…
Preview: OpenCode Full Tutorial: Free Models, Skills & MCPs
youtube
Plan/Build and persistent project instructions in AGENTS.md. Not independent rerun or known-answer tests; free-model claims are not current pricing/privacy evidence.
Matching text… free-model claims are not current pricing or privacy evidence.
T20-L05: See how to keep project instructions in AGENTS.md rather than repeating them in every OpenCode session. This does not…
Preview: Claude Managed Agents and Evolution of Agentic Surfaces — Gagan Bhat & Isabella Kai He, Anthropic
youtube
Distinguish model reasoning, the harness and tool execution. Technical vendor-specific companion, not comparison of the three book harnesses; no adoption of performance claims.
Matching textT20-L01: Hear why a model needs surrounding software to fetch information, use tools and carry out a task. This technical, vendor-specific conceptual companion is not a beginner setup guide or…
Preview: How to Add LM Studio AI Models to OpenCode for Coding
youtube
Connect OpenCode to LM Studio via custom provider, base URL and model ID. OpenCode/LM Studio only, not DeepSeek Harness, gateway security or complete tool-call acceptance; no blanket privacy claim.
Matching textT20-L03: Follow the OpenCode–LM Studio connection using a custom provider, base URL, /v1 endpoint and exact model ID. This September 2026 walkthrough covers that connection only—not DeepSeek…