flow-next-audit
Audit `.flow/memory/` entries against the current codebase and decide Keep / Update / Consolidate / Replace / Delete / Harden per entry. Triggers on…
Use while drafting any substantial reply, report, review walkthrough, or summary for the user - read the artifact prose contract before writing, not after, and draft under its rules. Do not use for short conversational turns, tool-call narration, the visual digest, or output
$ npx -y skills add gmickel/flow-next --skill flow-next-prose --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/flow-next-proseContext preview
The summary Claude sees to decide when to auto-load this skill.
Use while drafting any substantial reply, report, review walkthrough, or summary for the user - read the artifact prose contract before writing, not after, and draft under its rules. Do not use for short conversational turns, tool-call narration, the visual digest, or output
name: flow-next-prose description: "Use while drafting any substantial reply, report, review walkthrough, or summary for the user - read the artifact prose contract before writing, not after, and draft under its rules. Do not use for short conversational turns, tool-call narration, the visual digest, or output that lands in a file, PR, or tracker (those artifact surfaces carry their own pointers). Also invocable directly as /flow-next:prose [draft]." user-invocable: false allowed-tools: Read
Ambient triggering is host-discretionary on every host: the host matches this skill's description at the moment the agent begins drafting a substantial reply.
The agent reads the prose contract at [docs/prose.md](../../docs/prose.md), resolved relative to this SKILL.md file's own location. If the doc is absent, the skill proceeds without it and never blocks the reply.
The agent applies the contract's rules 1-10 at draft time and writes the reply under them from the first draft.
The reply-dormant parts of the doc are the precedence section's dedup-marker and projection bullets, which stay idle for replies. Rule 8 (user-outcome-first ordering) softens to lead with the answer. Rule 10 (honesty) stays fully active.
The agent never rewrites quoted material, code, or command output, and never restyles the user's words where they are being quoted back. A draft the user hands the skill as the `/flow-next:prose <draft>` argument is submitted for tightening, so the agent rewrites it under the contract.
Pass the identity (the path above). Never copy rule text from the doc into this skill.
Repeatable agentic engineering. The workflow layer that turns AI coding agents into a disciplined factory: durable specs, fresh-context workers, adversarial cross-model reviews, receipts. Everything in your repo, zero dependencies. Claude Code · Codex · Cursor · Droid.
Audit `.flow/memory/` entries against the current codebase and decide Keep / Update / Consolidate / Replace / Delete / Harden per entry. Triggers on…
Synthesize the current conversation context into a flow-next spec at `.flow/specs/<spec-id>.md` via `flowctl spec create + spec set-plan` — agent-native,…
Decision-map discovery for one oversized/unclear idea before capture. Triggers on /flow-next:chart with an unshaped idea, chart id, decision pin, --status, or…
Show spec dependency graph and execution order. Use when asking 'what's blocking what', 'execution order', 'dependency graph', 'what order should specs run',…
Drive any UI surface like a real user - a web app, a Chromium-backed desktop app (Electron / WebView2, reached over CDP), or a genuinely native app (macOS…
Export RepoPrompt context to a markdown file for review with an external LLM (ChatGPT, Claude web, etc.). Use when you want Carmack-level review but prefer an…