SpecStory: Turning AI Coding Chats into Reusable Skills
hugobowne · x · 2026-07-20
SpecStory saves sessions from coding agents like Claude Code and Cursor as Markdown, using Lore to mine for tacit knowledge within them.
The author demonstrates a complete workflow based on 516 sessions:
- Local Indexing: Avoids sending massive amounts of tokens to the model.
- Behavior Extraction: Transforms each interaction into evidence-backed "beats," determining whether the agent's work was accepted, rejected, or modified based on the user's subsequent actions.
- Skill Consolidation: Identifies recurring patterns across projects to generate skill profiles with references to original sessions, leaving it up to humans to decide whether to incorporate them into the skill library.
This solution effectively prevents the loss of decision-making processes after coding agent sessions are closed.
Related event: SpecStory turns AI coding chats into reusable skills(2 posts)→
More from coding & agent
- Building a Secure AI Agent Gateway: Self-Hosting OAuth for Multiple SaaS Apps — Defiant_Cod_2654 · 2026-07-22
- Rowboat launches as an open-source, local-first AI coworker with memory — ycombinator · 2026-07-22
- Reddit user chains Ideogram 4 and Krea2 to mimic bbox-based image positioning — v3lh0t05c0 · 2026-07-22
- Apollo Cuts AI Assistant Skill Dev Time by 85% with Deep Agents — LangChain · 2026-07-22
- Scoble says AI “loops” really means long-running multi-agent workspaces — Scobleizer · 2026-07-22
- Kimi Code opens a waitlist as Moonshot rolls out its coding product — Fabulous_Bonus_8981 · 2026-07-22