sefxvn16

Project Feasibility — Development skill for Claude Code

Development community

A Claude skill for preliminary project financial feasibility (NPV, IRR, payback, sensitivity) for new investments, M&A and partnerships.

How to install Project Feasibility

This entry records only its repository, not the path inside it, so there is no exact command to give. Open sefxvn16/project-feasibility and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Project Feasibility does

A Claude skill for preliminary project financial feasibility (NPV, IRR, payback, sensitivity) for new investments, M&A and partnerships. Enforces assumption discipline: Fact/Estimate/Assumption labels, a 3-phase pipeline with review stops, and Go/No-go conclusions. Skill text in Vietnamese

Alternatives in Development

  • Identify Assumptions Existing — Identify risky assumptions across Value, Usability, Viability, Feasibility 7.8k ★
  • Lbo — LBO 快速测试 — 假设 PE 买方用 5x 杠杆能否赚 20%+ IRR 6.6k ★
  • Academic Pptx Skill — A Claude Skill for creating academic presentations (conference talks, seminar slides, thesis defenses, grant b 811 ★

README

project-feasibility

**English** | [Tiếng Việt](README.vi.md)

A Claude skill for building and analyzing project financial feasibility models (cash flow, NPV, IRR, Payback, sensitivity analysis, investment conclusion) for **preliminary internal decisions** on a specific project: a new investment, an M&A deal or a business partnership. Its focus is assumption discipline: every input is labeled, the work moves through three phases with a stop after each, and assumptions cannot be adjusted to make the result look better.

The skill instructions (`SKILL.md`) are currently written in Vietnamese. Technical terms (NPV, IRR, Fact, Estimate, Assumption, F1–F3…) are kept in English.

What it does

A feasibility model can look reasonable at the end even when an early assumption is wrong, and there is a pull in the other direction too: quietly tuning inputs until the NPV looks right. The skill works against both with three rules.

**1. Every statement carries exactly one label**

  • Fact: sourced, observable or published.
  • Estimate: always given as a range, never a single number.
  • Assumption: an internal assumption, used only when no reliable Fact or Estimate exists.

**2. A one-way, three-phase pipeline with a stop after each phase**

Phase Does Does not
F1 — Data & Assumptions Labels inputs, rates their reliability, records gaps and out-of-model risks Calculate, evaluate, conclude
F2 — Modeling & Calculation Follows your file template, or proposes a file structure for approval first; builds a traceable model (inputs / formulas / outputs kept separate), calculates NPV, IRR, Payback Create or change assumptions, run sensitivity, interpret
F3 — Analysis & Decision Sensitivity, break-even, scenario comparison, conclusion Change inputs or formulas, drop the bad case, promise returns

After each phase the skill presents the result and waits for the user to confirm before moving on. Changing an assumption