返回技能列表

generate-status-report

pjt222
更新于 2 days ago
5 次查看
17
2
17
在 GitHub 上查看
data

关于

This Claude Skill automatically generates a project status report by reading project artifacts like the charter and backlog to calculate metrics and identify blockers. It summarizes progress using RAG indicators for schedule, scope, budget, and quality. Use it at the end of a sprint, before governance meetings, or when a new risk emerges.

快速安装

Claude Code

推荐
主要方式
npx skills add pjt222/agent-almanac -a claude-code
插件命令备选方式
/plugin add https://github.com/pjt222/agent-almanac
Git 克隆备选方式
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/generate-status-report

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

Generate a Project Status Report

Periodic status: read artifacts → metrics → summary (accomplishments, blockers, next) → RAG health (sched/scope/budget/quality).

Use When

  • End of sprint/period (wk/bi-wk/mo)
  • Stakeholder health req
  • Pre steering/governance meeting
  • Health change (new blocker/risk)
  • Checkpoint vs charter milestones

In

  • Required: period (start, end)
  • Required: ≥1 artifact (BACKLOG.md, SPRINT-PLAN.md, WBS.md, PROJECT-CHARTER.md)
  • Optional: prior reports (trend)
  • Optional: budget/resource data
  • Optional: risk register updates

Do

Step 1: Read artifacts

Scan proj dir:

  • PROJECT-CHARTER.md — milestones, success criteria
  • BACKLOG.md — item counts, burn-down
  • SPRINT-PLAN.md — goal, committed, task done
  • WBS.md — work pkg %
  • Prior STATUS-REPORT-*.md — trend

Adapt to what exists.

→ ≥1 read, key metrics extracted.

If err: no artifacts → can't generate. Create charter/backlog first via draft-project-charter or manage-backlog.

Step 2: Metrics

Agile (BACKLOG.md / SPRINT-PLAN.md):

  • Velocity: pts this sprint
  • Completion: done/committed
  • Burn-down: remaining vs prev
  • Cycle time: avg days In Progress → Done

Classic (WBS.md):

  • % complete: pkgs done / total
  • Schedule variance: planned vs actual
  • Effort variance: est vs consumed
## Metrics
| Metric | Value | Previous | Trend |
|--------|-------|----------|-------|
| Velocity | [N] pts | [N] pts | ↑/↓/→ |
| Sprint Completion | [N]% | [N]% | ↑/↓/→ |
| Backlog Remaining | [N] pts | [N] pts | ↓ (good) |
| Schedule Variance | [+/-N days] | [+/-N days] | |

→ 3-5 metrics w/ prev comparison.

If err: no historical (first) → omit prev + trend cols. Incomplete → note gaps + track action items.

Step 3: Blockers, risks, issues

## Blockers & Risks
| ID | Type | Description | Severity | Owner | Status | Action Required |
|----|------|------------|----------|-------|--------|----------------|
| R-001 | Risk | [Description] | High | [Name] | Open | [Action] |
| B-001 | Blocker | [Description] | Critical | [Name] | Active | [Action by date] |
| I-001 | Issue | [Description] | Medium | [Name] | Investigating | [Action] |

Cross-ref charter risk register. Flag new.

→ All blockers + top risks w/ owners + actions.

If err: no blockers → explicit "No active blockers", don't leave empty. No owner → escalate to PM.

Step 4: Accomplishments + next plan

## Accomplishments (This Period)
- [Completed item/milestone with evidence]
- [Completed item/milestone with evidence]
- [Completed item/milestone with evidence]

## Planned (Next Period)
- [Planned item/milestone with target]
- [Planned item/milestone with target]
- [Planned item/milestone with target]

→ 3-5 accomplishments w/ evidence, 3-5 planned.

If err: no accomplishments → reason (blocked, replanning, unavail). Next unclear → "Planning session scheduled for [date]".

Step 5: RAG + write

DimensionGreenAmberRed
ScheduleOn track or ahead1-2 weeks behind>2 weeks behind or milestone missed
ScopeNo uncontrolled changesMinor scope adjustmentsScope creep affecting deliverables
BudgetWithin 5% of plan5-15% over plan>15% over plan or untracked
QualityTests pass, criteria metMinor quality issuesCritical defects or acceptance failures

Write:

# Status Report: [Project Name]
## Report Date: [YYYY-MM-DD]
## Reporting Period: [Start] to [End]
## Document ID: SR-[PROJECT]-[YYYY-MM-DD]

### Overall Health
| Dimension | Status | Notes |
|-----------|--------|-------|
| Schedule | 🟢/🟡/🔴 | [One-line explanation] |
| Scope | 🟢/🟡/🔴 | [One-line explanation] |
| Budget | 🟢/🟡/🔴 | [One-line explanation] |
| Quality | 🟢/🟡/🔴 | [One-line explanation] |

### Executive Summary
[2-3 sentences: overall status, key achievement, biggest risk]

### Metrics
[From Step 2]

### Accomplishments
[From Step 4]

### Blockers & Risks
[From Step 3]

### Planned Next Period
[From Step 4]

### Decisions Needed
- [Decision 1 — needed by date, from whom]

---
*Report prepared by: [Name/Agent]*

Save STATUS-REPORT-[YYYY-MM-DD].md.

→ Complete report w/ RAG + metrics + narrative.

If err: insufficient for RAG → ⚪ (Grey) "insufficient data" + list data needed next report.

Check

  • Report file created, date-stamped
  • RAG all 4 dims w/ justify
  • ≥3 metrics from artifacts
  • Blockers section present (even "No active")
  • Accomplishments w/ evidence
  • Next plan
  • Exec summary 2-3 sentences
  • Every blocker/risk has owner + action + deadline

Traps

  • No data: evidence-based. Every claim → artifact/metric.
  • All green always: no evidence = not honest. Challenge green.
  • Blocker w/o owner: no owner → not resolved.
  • Metric w/o context: "Velocity=18" meaningless alone. Always prev/target.
  • Too long: 2 min scan. 1-2 pages.
  • Missing decisions: explicit w/ deadlines.
  • Stale data: verify artifact dates match period.
  • Missing trends: first-time OK, later must compare.

  • draft-project-charter — milestones + success criteria
  • manage-backlog — backlog metrics feed report
  • plan-sprint — sprint results: velocity + completion
  • create-work-breakdown-structure — WBS % drives classic metrics
  • conduct-retrospective — report data feeds retro

GitHub 仓库

pjt222/agent-almanac
路径: i18n/caveman-ultra/skills/generate-status-report
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

相关推荐技能

content-collections

Content Collections 是一个 TypeScript 优先的构建工具,可将本地 Markdown/MDX 文件转换为类型安全的数据集合。它专为构建博客、文档站和内容密集型 Vite+React 应用而设计,提供基于 Zod 的自动模式验证。该工具涵盖从 Vite 插件配置、MDX 编译到生产环境部署的完整工作流。

查看技能

polymarket

这个Claude Skill为开发者提供完整的Polymarket预测市场开发支持,涵盖API调用、交易执行和市场数据分析。关键特性包括实时WebSocket数据流,可监控实时交易、订单和市场动态。开发者可用它构建预测市场应用、实施交易策略并集成实时市场预测功能。

查看技能

creating-opencode-plugins

该Skill帮助开发者创建OpenCode插件,用于接入命令、文件、LSP等25+种事件。它提供了插件结构、事件API规范和JavaScript/TypeScript实现模式,适合需要拦截操作、扩展功能或自定义事件处理的场景。开发者可通过它快速构建响应式模块来增强OpenCode AI助手的能力。

查看技能

sglang

SGLang是一个专为LLM设计的高性能推理框架,特别适用于需要结构化输出的场景。它通过RadixAttention前缀缓存技术,在处理JSON、正则表达式、工具调用等具有重复前缀的复杂工作流时,能实现极速生成。如果你正在构建智能体或多轮对话系统,并追求远超vLLM的推理性能,SGLang是理想选择。

查看技能