flow-next-capture

Synthesize conversations into Flow-Next epic specs with source-tagged evidence.

674|52|Updated Dec 26, 2025
One-click install
npx skills add https://github.com/gmickel/flow-next --skill flow-next-capture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: flow-next-capture
Source: https://github.com/gmickel/flow-next/tree/main/plugins/flow-next/codex/skills/flow-next-capture
Command: npx skills add https://github.com/gmickel/flow-next --skill flow-next-capture

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Transforms a live conversation into a structured Flow-Next epic spec, preserving exact user input quotes and agent inferences to enable reliable planning and auditing.

Core Features & Use Cases

  • Source-tagged, per-line evidence in the resulting epic (user quotes, paraphrases, and inferred content).
  • Supported in interactive and autofix modes, with duplicate detection, memory context referencing, and a separate read-back phase prior to write.
  • Generates tactical guidance for next steps and plan-wide handoffs to /flow-next:plan and /flow-next:interview.

Quick Start

Initiate a capture session to convert the current conversation into a Flow-Next epic spec saved at .flow/specs/<epic-id>.md.

Frequently Asked Questions about flow-next-capture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert a conversation into an epic spec with source-tagged evidence?

To convert a conversation into an epic spec, this Skill synthesizes your current chat into a Flow-Next spec file using source-tagged, per-line evidence, preserving exact user quotes and agent inferences.

How do I create an epic spec from a live conversation without losing context?

Creating an epic spec from a live conversation requires a mandatory read-back phase before writing the file, ensuring all user quotes and paraphrases are verified and context is accurately preserved at .flow/specs/<epic-id>.md.

Does the epic spec capture process support automated duplicate detection?

Yes, the epic spec capture process supports automated duplicate detection alongside memory context referencing, operating in both interactive and autofix modes to handle overlapping content during synthesis.

Can I use autofix mode to rewrite compacted conversations into epic specs?

You can use autofix mode to rewrite compacted conversations by enabling the optional --rewrite and --from-compacted-ok controls, allowing automated handling of memory context hits and duplicates during spec generation.

What are the limitations of automated conversation capture for epic specs?

A key limitation of automated conversation capture is the strict mandatory read-back requirement prior to writing, meaning the synthesis process cannot skip user verification even when running in autofix mode.

What should I do after generating a Flow-Next epic spec file?

After generating a Flow-Next epic spec, you should follow the generated next-step footer guidance, which directs you to tactical planning and detailed interviewing using /flow-next:plan and /flow-next:interview.