harness-engineering
Harness Engineering
Design environments, feedback loops, and control systems that enable AI coding agents to produce reliable software at scale. A harness is everything around the code that keeps agents effective: structured knowledge, mechanical constraints, application legibility, and entropy management.
Core insight: When agents struggle, the fix is almost never "try harder." Ask: what capability is missing, and how do we make it both legible and enforceable for the agent?
Variant: harness-engineering-copilot — strategies and patterns specialized for GitHub Copilot agent customization.
Sources: OpenAI — Harness Engineering | Martin Fowler — Harness Engineering
Three Harness Components
| Component | What it does | Deterministic? |
|---|---|---|
| Context Engineering | Curated, structured knowledge base in the repo + dynamic context (observability, browser) | Mixed |
| Architectural Constraints | Layered architecture enforced by custom linters and structural tests | Deterministic |
| Garbage Collection | Periodic agents that detect drift, stale docs, and constraint violations | LLM-based |
Assess Current Harness
More from arisng/github-copilot-fc
skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
20openspec-sdd
>-
2blazor-blueprint-ui
Build and customize .NET 8+ Blazor UIs with BlazorBlueprint. Use when choosing between BlazorBlueprint.Components and BlazorBlueprint.Primitives, wiring setup and providers, using ToastService/DialogService/localization, selecting icon packs, applying shadcn-ui-style theming, or copying and adapting BlazorBlueprint blueprints.
2beads
Track complex, multi-session work with dependency graphs using beads issue tracker. Use when work spans multiple sessions, has complex dependencies, or requires persistent context across compaction cycles. For simple single-session linear tasks, TodoWrite remains appropriate.
1ralph-knowledge-merge-and-promotion
Ralph-v2 knowledge extraction, staging, merge, and promotion reference. Use when running Librarian EXTRACT, STAGE, or PROMOTE modes, applying merge rules, writing extracted frontmatter, or promoting staged knowledge into `.docs`.
1agent-evaluator
Deterministic custom subagent selection helper. Use when you need a reproducible, auditable decision on which custom subagents to activate for a user query (runs scripts/agent_evaluator.py).
1