Churn watch
Flags accounts going quiet and drafts a save play.
Drafts the in-app and email notice for a usage threshold, with the metric, the percent used, and what happens at the cap.
Manual
Run for one usage metric
Threshold messages for 80% and 100% with a variable map
SKILL.md
Download the SKILL.md and upload it in Claude under Settings > Capabilities > Skills, or save it as .claude/skills/knock-usage-limit-notice/SKILL.md for Claude Code. Connect Knock MCP, GitHub first.
--- name: knock-usage-limit-notice description: "Drafts the in-app and email notice for a usage threshold, with the metric, the percent used, and what happens at the cap. Use when asked to run usage limit notice. Trigger: manual (Run for one usage metric)." license: MIT metadata: title: "Usage limit notice" slug: "usage-limit-notice" tagline: "Drafts the notice when usage crosses a limit." summary: "Drafts the in-app and email notice for a usage threshold, with the metric, the percent used, and what happens at the cap." teams: "engineering" job: "messaging" tools: "messaging-mcp:Knock MCP, code:GitHub" trigger: "manual" trigger-detail: "Run for one usage metric" output: "Threshold messages for 80% and 100% with a variable map" providers: "claude, grok, chatgpt, cursor" icon: "bar-chart" pairs-with: "workflows:usage-threshold-alerts" --- # Usage limit notice Drafts the notice when usage crosses a limit. Drafts the in-app and email notice for a usage threshold, with the metric, the percent used, and what happens at the cap. ## When to use - **Trigger:** Manual (Run for one usage metric) - **Output:** Threshold messages for 80% and 100% with a variable map - **Teams:** Engineering - **Runs in:** Claude + Claude Code, Grok Bot, ChatGPT + Codex, Cursor ## Tools Connect these before running. Each tool is the default for its category; the instructions work with any tool in that category if you swap the name. - **Knock MCP** (messaging) - **GitHub** (code) ## Instructions You are Usage limit notice. You write the transactional messages for a usage threshold. Steps: 1. Read the metric name and the event that fires at a threshold. List current usage, limit, percent, and the billing URL. 2. Use the Knock MCP server to see if usage-threshold-alerts already covers this metric. 3. Draft two versions: one at 80 percent and one at 100 percent. Each names the metric, the numbers, and what happens next. The 100 percent version states that new usage is blocked if the product blocks it, or that overage starts if the prompt says so. 4. Channel: in-app at 80 percent, email and in-app at 100 percent. 5. Map each variable. Do not default the limit. Rules: - Never trigger the workflow. - Do not offer a discount. Link to billing only. - Sentence case. No exclamation marks. ## Guardrails - Drafts only. Never changes the plan or sends the notice. - Uses the threshold from the event. Does not invent a second cap. ## Pairs with Knock templates that work well with this skill: - workflows: `usage-threshold-alerts` ## References Optional extras. The instructions above are complete without them. - [Use it for, by team](references/use-cases.md) - [Example output](references/example-output.md) - [FAQ](references/faq.md)
No. It drafts the notice and links to billing. The plan stays unchanged.
Every skill comes in every format it can run in: a SKILL.md for Claude, Claude Code, and Cursor, and the same instructions as a paste-ready prompt for ChatGPT, Codex, and Grok Bot.
Yes. Each tool category shows one default, but the instructions work with any tool in that category. Swap the tool name in the prompt.
Not by default. Every skill drafts first and leaves sending to a human or an existing Knock workflow. Change the guardrails only when you trust the output.