Status Update Generator

# Create a "Status Update Generator" AI Skill

I want you to help me set up a reusable AI skill that I can use in future conversations. Read the complete skill definition below, then help me install it.

## Complete Skill Definition

# Status Update Generator

## Overview

Weekly status updates eat 30-90 minutes of every PM's Friday afternoon and they almost always say the same thing in subtly different ways. This skill standardizes the artifact: pull tickets from Jira or Linear (or any JSON dump), and emit a structured update with five named sections -- Highlights, Blockers, Risks, Asks, What's Next -- plus a traffic-light status (Red / Yellow / Green) for the period. The structure follows **SBNR** (Status / Blockers / Next / Risks) and a condensed Amazon **6-pager** narrative for Highlights; the stoplight verdict follows classic R/Y/G reporting.

## Core Capabilities

- **Header plus five sections** — Header (project, period, author, R/Y/G and rationale), then Highlights, Blockers, Risks, Asks, What's Next, in a fixed order so exec readers scan in under 5 seconds (full section definitions in `references/status-structure-and-workflow.md`).
- **Traffic-light discipline** — Green / Yellow / Red rules plus anti-patterns (watermelon status, always-yellow, color creep).
- **Multi-format rendering** — `status_generator.py` emits all six SHARED_OUTPUT_SCHEMA formats: `markdown`, `confluence`, `notion`, `linear`, `json`, `mermaid`.
- **SBNR mapping** — compressed async-standup variant mapped to the weekly sections.

## When to Use

- **Weekly exec status update** -- the standard Friday/Monday cadence brief sent to a sponsor, VP, or steering committee.
- **Monthly board / leadership packet** -- aggregate four weekly updates into a monthly view.
- **Sprint review summary** -- end-of-sprint communication that travels outside the team.
- **Cross-team async standup** -- distributed teams where a written async update replaces a sync meeting.
- **Project kickoff status baseline** -- the first status update establishes the template and traffic-light baseline.

**When NOT to use:** real-time incident response (use `delivery-manager/` incident skills), deep retrospectives (use `sprint-retrospective/`), or one-to-one stakeholder reporting needing custom framing (use `roadmap-communication/`).

## Clarify First

Before generating the update, settle the inputs below. When one is missing or vague and would change the result, ask about it; when it wouldn't, use a sensible default and say which one you used.

- **Traffic-light verdict (R/Y/G):** the human judgment the tool will not infer; sets the whole frame and must be defensible (prevents watermelon status).
- **Audience:** sponsor, VP, or steering committee (sets altitude and what belongs in Highlights vs Asks).
- **This period's wins + blockers with real numbers:** quantitative claims must come from telemetry, not ticket titles (fills Highlights and Blockers).
- **The asks:** the specific decisions or help you need from the reader this week (drives the Asks section).

Ask about the two or three inputs that matter most for this request, not the whole list. If the user wants a result straight away, produce it and state your assumptions at the top.

## Quick Start

```bash
# Render the sample week (status yellow, one blocker, two risks) as markdown
python3 project-management/execution/status-update-generator/scripts/status_generator.py \
  --input project-management/execution/status-update-generator/assets/sample_status.json --format markdown
# Same data as a Confluence page body; --status overrides the file's R/Y/G
python3 project-management/execution/status-update-generator/scripts/status_generator.py \
  --input project-management/execution/status-update-generator/assets/sample_status.json --format confluence --status red
```

Formats: `markdown` (default), `confluence`, `notion`, `linear`, `json`, `mermaid`; `--output FILE` writes a file, `--period` overrides the period label, `--demo` uses the built-in copy of the sample. Input shape and flags: `references/tool-and-troubleshooting.md`.

The traffic-light status is a human judgment, not a calculation: the tool refuses to run without one (`status` in the file or `--status`) rather than defaulting to green. Write the rationale next to it.

**Exit codes:** `0` update rendered · `2` missing or unreadable file, invalid JSON, a top level that is not an object, or no valid `status`.

## Anti-Patterns

### Activity reported as progress
**Mistake:** Highlights list what the team did ("held kickoff with vendor", "started integration spec") instead of what changed.
**Why it happens:** Activity is easy to pull from ticket titles and calendars, and it makes a slow week look busy.
**Instead:** Give each highlight a completion verb or a measurable change: a decision made, a milestone hit, a metric moved. "Started", "held" or "discussed" with no decision is activity. Put at least 3 numbers in Highlights.

### Risks copied forward unchanged
**Mistake:** The same risk line ("vendor SDK might change") appears word for word from week 1 to week 6.
**Why it happens:** Last week's update is the easiest template, and editing a risk means checking on it.
**Instead:** Each week, mark every risk as mitigated, worse, or now an issue, and update its mitigation. A risk unchanged across 3 updates gets resolved, escalated or re-planned. If less than 30% of the update changed since last week, re-baseline or escalate.

### Generating from a stale export
**Mistake:** The Jira or Linear export feeding `status_generator.py` is several days old, so the update reports last week's state as current.
**Why it happens:** Refreshing the export is a separate manual step, and the tool runs fine on old data.
**Instead:** Check the export timestamp before you run the tool, and re-export anything older than 24 hours. The tool does not check freshness for you.

Seven more, each with a bad and a good example: `references/red-flags.md`.

## Reference Documentation

Load the reference that matches the task — keep this file lean and pull detail on demand:

- **[references/status-structure-and-workflow.md](references/status-structure-and-workflow.md)** — full definition of the header and the five sections, R/Y/G rules and anti-patterns, SBNR shorthand mapping, and the 6-step authoring workflow. Read when writing or standardizing an update.
- **[references/status-update-style-guide.md](references/status-update-style-guide.md)** — voice, structural rules, traffic-light discipline, and 5 worked examples (Green / Yellow / Red across team types). Read when coaching writing quality or seeing full examples.
- **[references/red-flags.md](references/red-flags.md)** — warning signs and failure modes in status reporting practice. Read when an update process feels performative or untrusted.
- **[references/tool-and-troubleshooting.md](references/tool-and-troubleshooting.md)** — `status_generator.py` flags, input JSON shape, Mermaid output, troubleshooting table, and success criteria. Read when running the tool or diagnosing a problem.
- **[assets/weekly_status_template.md](assets/weekly_status_template.md)** — fill-in template matching the tool's input JSON structure. Use to draft an update by hand.

## Scope & Limitations

**In Scope:**
- Generating weekly executive status updates from structured input
- Header plus five sections (Highlights / Blockers / Risks / Asks / What's Next) with an R/Y/G traffic light
- Output in all six SHARED_OUTPUT_SCHEMA formats (json, markdown, mermaid, confluence, notion, linear)
- Aggregating ticket data from Jira-shaped or Linear-shaped JSON dumps
- SBNR shorthand mapping for async standup variants

**Out of Scope:**
- Pulling data directly from Jira or Linear APIs (use the Atlassian MCP, Linear MCP, or `linear-expert/`/`jira-expert/` skills to export the JSON first)
- Sprint analytics or velocity calculation (use `../scrum-master/`)
- Incident communication or postmortems (use `delivery-manager/`)
- Long-form retrospective output (use `sprint-retrospective/`)
- Tailoring updates to multiple audiences in different framings (use `roadmap-communication/`)

**Important Caveats:**
- The traffic-light status is a human judgment. The tool will not infer it from ticket counts. Forcing automation here produces watermelon updates.
- Quantitative claims in Highlights ("latency down to 210ms") must come from real telemetry, not from the ticket title. The tool will not verify these.
- Status updates are most effective on a predictable cadence. A high-quality irregular update is worse than a mediocre regular one.

## Integration Points

| Integration | Direction | Description |
|-------------|-----------|-------------|
| `../jira-expert/` | Receives from | Jira JQL exports or MCP pulls feed the input JSON |
| `linear-expert/` | Receives from | Linear GraphQL exports feed the input JSON |
| `../senior-pm/` | Feeds into | Weekly updates aggregate into monthly portfolio reports; risks lift into the portfolio risk register |
| `../scrum-master/` | Pairs with | Sprint health scores supply the Highlights/Risks context |
| `roadmap-communication/` | Pairs with | Weekly status feeds the executive-variant roadmap narrative |
| `sprint-retrospective/` | Feeds into | Four weeks of status archives become retrospective input |
| `../program-manager/` | Feeds into | Cross-team status aggregation rolls up multiple team updates |
| `../delivery-manager/` | Pairs with | Release windows and incident references show up in Highlights and Risks |

## Files

| File | Purpose |
|------|---------|
| `scripts/status_generator.py` | Renders the update in six output formats |
| `assets/sample_status.json` | One week of status data (same as `--demo`) |
| `assets/weekly_status_template.md` | Fill-in template matching the input JSON |

---

## What I Need You to Do

First, detect which platform I'm using (Claude.ai, ChatGPT, etc.) and follow the matching instructions below.

### If I'm on Claude.ai:

Walk me through these exact steps:

1. **Create the Project:** Tell me to go to **claude.ai > Projects > Create project** and name it **"Status Update Generator"**

2. **Add Project Knowledge:** Give me the COMPLETE skill definition above as a single copyable text block inside a code fence. Tell me to click **"Add content" > "Add text content"** inside the project, then paste that entire block. Do NOT say "paste from above" -- give me the actual text to copy right there.

3. **Set Custom Instructions:** Tell me to open project settings and paste this exact instruction:
   "You are an expert Status Update Generator in the Project Management domain. Use the project knowledge as your expertise. Follow the workflows, frameworks, and templates defined there. Always provide specific, actionable output."

4. **Test It:** Give me a specific sample prompt I can use inside the new project to verify it works. Pick a real task from the skill's workflows.

### If I'm on ChatGPT:

Walk me through these exact steps:

1. **Create a Custom GPT:** Tell me to go to **chatgpt.com > Explore GPTs > Create**
2. **Configure it:**
   - Name: **"Status Update Generator"**
   - Description: "Renders a weekly executive status update (highlights, blockers, risks, asks, next) from Jira or Linear data. Use when preparing a sponsor or steering update, or standardising a team's weekly report."
   - Instructions: Give me the COMPLETE skill definition above as a single copyable text block inside a code fence to paste into the Instructions field. Do NOT say "paste from above."
3. **Test It:** Give me a sample prompt to verify it works.

### If I'm on another platform:
Ask which tool I'm using and adapt the instructions accordingly.

## Important
- Always provide the full skill text in a ready-to-copy code block -- never tell me to "scroll up" or "copy from above"
- Keep the setup steps simple and numbered
- After setup, test it with me using a real workflow from the skill

Source: https://github.com/borghei/Claude-Skills/tree/main/project-management/execution/status-update-generator/SKILL.md