agent-harness-orchestration
Agent Harness Orchestration
Practical workflow for running many coding agents. Humans steer. Agents execute. Harness supplies context, tools, guardrails, and feedback loops.
What It Is
A control system for coding agents. It turns tasks into isolated workspaces, keeps agents moving, verifies results, and consolidates changes without corrupting shared state.
Use repository-local knowledge as source of truth. Use issues, plans, or task lists as control plane. Use tests, linters, CI, app automation, logs, and review loops as feedback.
When to Use
- Multiple agents work on same repo
More from flpbalada/fb-skills
progressive-disclosure
Reduce complexity by revealing information progressively. Use when designing
10discuss-task
Clarify ambiguous tasks before action. Use when goal, scope, success criteria, constraints, or risks are unclear.
7react-useeffect-avoid
Guides when NOT to use useEffect and suggests better alternatives. Use when reviewing React code, troubleshooting performance, or considering useEffect for derived state or form resets.
6discuss-code
Critically discuss code issues with compact findings. Use when code needs review for logic, simplicity, structure, naming, or maintainability.
6react-key-prop
Guides proper usage of the key prop in React lists. Use this skill when rendering lists, mapping arrays to components, or troubleshooting list-related state bugs.
5cognitive-fluency-psychology
Apply cognitive fluency principles to improve clarity, trust, and conversion.
5