AC harness
Agent-first engineering knowledge base distilled from OpenAI's Harness Engineering post. Use this skill when the user asks about: agent-first development, agentic workflows, how to structure codebases for AI agents, context management for agents, AGENTS.md best practices, agentic code review, autonomous coding agents, agent-driven architecture, Codex engineering, how to work with AI coding agents, agent engineering principles, progressive disclosure in docs, architecture constraints for agents, error tolerance strategy, technical debt garbage collection, agent-readable observability, self-review loops, merge strategy for high-throughput agent systems, 智能体工程, 智能体优先开发, 智能体驱动编程, AGENTS.md 设计, 上下文管理, 智能体自主性, 代码库对智能体可读, harness engineering, 渐进式披露, 架构约束早期化, 错误容忍策略, 技术债务垃圾回收, 合并策略, 智能体自我审查, Ralph Wiggum 循环, 智能体可观测性.
As a process C 52/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 9. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "source" - note
frontmatter-keyunknown frontmatter key "created"
Process rating: all ten parameters 52/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 1 mutating operations with no state check
- 40Consistency. Frontmatter name (harness) differs from the folder (skill-factory-claw)
- 70When it triggers. States when to use, but not when not to
- 100Tools and files. No external tools needed
- 100Steps. 18 steps
- 100Execution cost. Instruction body is 468 tokens
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 821: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +4Structure: 6 headings
- +3Step-by-step instructions: 18 items
- +4Has examples (1 code blocks)
- +4Reference files are cited in the instructions (3 of 3)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 84.