A practical checklist for turning company workflows into reusable Agent Skills
FinanceYF5 · x · 2026-11-15
A checklist for turning AI agents from hallucinating, start-from-scratch tools into reliable executors:
- Knowledge baseline: create a facts file of real company info the agent may cite; a style file with 50 great examples, 20 negative cases and banned words; keep everything versioned on GitHub.
- Unified rules: define task rules via CLAUDE.md or AGENTS.md; add cross-session memory with claude-mem or mem0.
- Proprietary data: pipe ads, CRM and payment data through Airbyte + ClickHouse, and keep accumulating data competitors can't get.
- Workflows to Skills: rank all company workflows by revenue created/saved, run each one manually while narrating a screen recording, then have the agent convert the recording into a Skill with steps, checkpoints and common errors.
- Recommended mature skill repos: anthropics/skills, obra/superpowers, mattpocock/skills, addyosmani/agent-skills; for marketing see marketingskills, Graphed Skills and Boring Marketing.
Related event: Greg Isenberg shares 49 rules for building an AI-native company(12 posts)→
More from coding & agent
- 159k Lovable Apps Scanned: Many Expose Chats and Passport Data via Misconfigured Supabase APIs — bibryam · 2026-11-15
- Rust + wgpu rewrite renders 250k-file dependency graph at 60fps on M3 — IgorCarron · 2026-11-15
- Agents Orchestration Is Mostly Token Furnaces Now, Argues Dev — pvncher · 2026-11-15
- DataTalksClub Founder Launches Full-Time AI Academy in Germany, Tuition State-Funded — Al_Grigor · 2026-11-15
- How far can a company take AI? Greg Isenberg's 49-point playbook — FinanceYF5 · 2026-11-15
- TensorFold PR fixes MCP servers blocked by macOS sandbox over dyld framework loading — AIFlow_ML · 2026-11-15