Skip to content

Project Estimator

Verified

Breaks down projects into tasks, estimates effort with confidence intervals, identifies dependencies and risks, and produces planning spreadsheets.

By Anthropic 5,900 v1.1.0 Updated 2026-03-10

Install

Claude Code

Copy the SKILL.md file to your project's .claude/skills/ directory

About This Skill

Project Estimator applies structured estimation techniques to combat the planning fallacy — the universal human tendency to underestimate how long things take. It produces estimates with honest uncertainty ranges, not false precision.

Estimation Methodology

Three-Point Estimation (PERT) For each task, generates three scenarios: - **Optimistic (O)** — everything goes right, no blockers - **Most Likely (M)** — realistic with typical interruptions - **Pessimistic (P)** — significant unknowns or complications

Weighted average: `(O + 4M + P) / 6` with standard deviation `(P - O) / 6`.

Work Breakdown Structure Decomposes the project top-down: - Epic → Feature → Story → Task - Each leaf-level task estimated independently - No single task estimated over 3 days (decompose further if needed)

Confidence Bands Aggregates task estimates into project-level confidence bands: - P50 (50% confidence deadline) - P80 (80% confidence deadline) - P95 (95% confidence deadline)

Delivered with the message: "We're 80% confident this will be done by [date]."

Risk Register

  • Identifies risks alongside estimates:
  • Technical risks (unfamiliar technology, external API dependencies)
  • Resource risks (key person dependency, part-time availability)
  • Scope risks (ambiguous requirements, stakeholder alignment gaps)

Each risk scored by probability × impact with mitigation suggestions.

Output Formats

  • CSV for import into Jira, Linear, or Notion
  • Markdown task list with time estimates
  • Gantt chart description for visualization tools

Use Cases

  • Estimating development sprints and releases for engineering teams
  • Creating project proposals with realistic timelines for clients
  • Breaking down product features into sized user stories
  • Building resource allocation plans for multi-team projects

Pros & Cons

Pros

  • + Three-point PERT estimation provides honest uncertainty ranges
  • + P50/P80/P95 confidence bands for stakeholder communication
  • + Risk register generated alongside estimates
  • + CSV output importable to Jira, Linear, and Notion

Cons

  • - Estimate quality depends on the detail of the project description provided
  • - Does not integrate with actual velocity data from your team's sprint history

Related AI Tools

Related Skills

Stay Updated on Agent Skills

Get weekly curated skills + safety alerts

每周精选 Skills + 安全预警