# Agent Harness Engineer

> Audit, redesign, and maintain the harness around an AI agent — the instruction, memory, tool, and check layer that shapes every run. Use when the user says "my agent keeps forgetting", "my instructions file is a mess", "audit my CLAUDE.md / AGENTS.md", "the agent ignores my rules", "context keeps getting re-derived every run", "set up my agent's operating files", or asks about agent memory, agent handoffs, SOUL.md, skill libraries, instruction bloat, harness design, or migrating agent context between tools. Covers mapping every instruction file, evidence-labeling rules, the six-outcome triage (keep / one-home / load-later / turn-into-check / probation / retire), the four-part core brief, knowledge-layer design with retrieval contracts, handoff records and receipts for multi-agent loops, SOUL.md elicitation interviews, 3x3 compact-vs-full experiments, and context portability audits. Trigger on any complaint that an agent's behavior drifted after a model upgrade or that nobody knows which rule wins.

## Facts
- Page: https://tashan.sh/capability/skill-cloud-byte-consulting-agent-harness-engineer
- tashan id: skill:Cloud-Byte-Consulting/agent-harness-engineer
- Source: https://github.com/Cloud-Byte-Consulting/plugins
- Type: skill
- Category: other
- tashan score: not scored (catalogued only — too little public evidence)
- Adoption: 9.0
- Upkeep: 95.0
- Freshness: 90.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- License: Apache-2.0
- Official: no

## Install

```sh
cp -r agent-harness-engineer ~/.claude/skills/
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-08-22 by tashan (https://tashan.sh) from public evidence. Scorer s5.