General skillAI and agent development
harness-check
Audit whether a multi-agent harness works as wired
- Stars
- 0
- Forks
- 0
- Updated
- Updated Jul 3, 2026
MIT
Audit whether a multi-agent harness works as wired
harness-check audits multi-agent frameworks assembled from prompts, Markdown, and scripts. It checks four dimensions: system architecture, trigger reliability, execution reliability, and declared-versus-wired behavior, covering 24 failure patterns involving discovery, registries, symlinks, script seams, locks, and state handling. Findings must be verified against real files, then appear in a behavior-focused report with severity, fix risk, and a split between safe fixes and decisions that need the owner’s call.
Resource types
Use cases
Platforms
Turn detailed business processes into reusable, tested AI skills.
Define portable task-specific sub-agents in Markdown for several coding assistants.
Capabilities
Audience
Public GitHub facts last synced Jul 10, 2026.
Find, create, run, and improve agent skills from real execution feedback.
Reduce LLM token use by pruning context, caching prompts, and routing models.