ScobleClaw monorepo
a9e0031721
Analyzes coach issues #24 (trust), #25 (engineering), and #26 (vault) — all scored 5/10 for repetitive behavior patterns. Identifies common root cause: all agents share deepseek-v4-flash:cloud model with no loop-detection guardrails or iteration budgets. Key findings: - All 3 critiques are VALID — real systemic issue, not coach noise - Root cause: single shallow model across all agents + no loop-breaker skills - Coach uses same model to critique itself (self-referential grading) - Three distinct failure patterns: yield loops, planning loops, cron loops - Recommended fixes: guardrail skills, staleness detection, model tiering Closes #9 |
||
|---|---|---|
| coach-pattern-analysis.md | ||
| README.md | ||
ScobleClaw
Agentic engineering framework by Scott Scoble — tools, workflows, and skills for building with AI agents.
What Is ScobleClaw?
ScobleClaw is a monorepo for agentic engineering patterns used across Scott Scoble's open-source projects. It provides:
- Subagent dispatch patterns for structured AI agent workflows
- Skill-based development with reusable, composable AI context modules
- Provenance tracking for auditable agent outputs and decisions
- Cross-repo tooling shared across Speckl, brand-site, beyond-the-prompt, and more
Projects
| Project | Description | Author |
|---|---|---|
| Speckl | Spec language compiler with embedded provenance | Scott Scoble |
| Brand Site | Professional speaker website | Scott Scoble |
| Beyond the Prompt | Workshop: Lessons from building with Claude Code | Scott Scoble |
Author
Created and maintained by Scott Scoble — builder, speaker, and AI agentics engineer.
- Website: scott.scoble.me
- LinkedIn: scottscoble
- GitHub: sscoble
License
MIT © Scott Scoble / Greybeard Holdings, LLC