How do I install the Tc Review skill?

Run npx skillmds@latest add hangnhat57/tc-review in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Tc Review safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: executes scripts, makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Tc Review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Tc Review free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Tc Review?

hangnhat57 (@hangnhat57) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Tc Review by hangnhat57 — Agent Skill

Name: Tc Review
Rating: 4.2 (5 reviews)
Author: hangnhat57

Tc Review

Review QA test case documents (Markdown) against source code. Use when: (1) Reviewing test cases for coverage gaps, (2) Learning source code to extract test scenarios, (3) Suggesting missing test cases, (4) Measuring test case quality. Supports multi-repo projects with dual memory (local MD + Dify RAG).

hangnhat57 Updated Sep 10, 2026

File contents

Test Case Review Agent

AI-powered test case document reviewer. Learns source code, then reviews test cases to find coverage gaps.

Quick Start

# Create project and add repo
tc-review project create myproject
tc-review project add-repo myproject https://github.com/org/repo.git

# Learn source code
tc-review learn myproject

# Review test cases
tc-review review tests/checkout.md --project myproject

Core Commands

Project Management

tc-review project create <name>           # Create project
tc-review project list                    # List projects
tc-review project show <name>             # Show details
tc-review project add-repo <name> <url>   # Add GitHub repo
tc-review project delete <name>           # Delete project

Learning & Review

tc-review learn <project>                 # Learn from all repos
tc-review review <file.md> --project <p>  # Review test case file
tc-review suggest <project>               # Suggest new test cases

Memory

tc-review memory status                   # Check local + Dify status
tc-review memory search <proj> <query>    # Search knowledge
tc-review memory sync <project>           # Sync to Dify RAG

Config

tc-review config show                     # Show current config
tc-review config init                     # Create config.json
tc-review config env                      # List env variables

LLM Configuration

Anthropic Claude (Default)

export LLM_PROVIDER=anthropic
export ANTHROPIC_API_KEY=sk-ant-...
export LLM_MODEL=claude-sonnet-4-20250514

OpenAI / Codex

export LLM_PROVIDER=openai
export OPENAI_API_KEY=sk-...
export LLM_MODEL=gpt-4-turbo

Test Case Format

Test cases must be in Markdown format:

# Test Suite: Checkout Flow

## TC-001: Successful Payment
**Preconditions:**
- User logged in
- Cart has items

**Steps:**
1. Navigate to checkout
2. Enter valid card
3. Click Pay

**Expected Result:**
Payment success, order created

Output Example

📊 Review Report: checkout.md
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Score: 65/100    Coverage: 65%

✅ Covered:
  • Successful payment
  • Invalid card

⚠️ Missing:
  • Payment timeout
  • Expired card
  • Concurrent payments

💡 Suggested:
  TC-003: Payment Timeout
  TC-004: Expired Card Validation

Workflows

Initial Setup

Create project: tc-review project create <name>
Add repos: tc-review project add-repo <name> <url>
Learn: tc-review learn <name>
(Optional) Sync to Dify: tc-review memory sync <name>

Review Cycle

QA writes test cases in Markdown
Run: tc-review review <file> --project <name>
Address gaps found
Re-review until score > 80

Get Suggestions

tc-review suggest myproject

Generates test case suggestions based on learned knowledge.

Memory System

Layer	Storage	Search	Best For
Local	MD + JSON	Keyword	Offline, fast
Remote	Dify RAG	Semantic	Large codebases

Dify RAG Setup (Optional)

export DIFY_ENABLED=true
export DIFY_API_KEY=app-...
export DIFY_DATASET_ID=...
export DIFY_BASE_URL=https://api.dify.ai/v1

Troubleshooting

No ANTHROPIC_API_KEY: Set export ANTHROPIC_API_KEY=sk-ant-...

Project not found: Run tc-review project list to see available projects

Low coverage score: Check gaps in report, add missing test cases

Dify not connected: Verify DIFY_API_KEY and DIFY_DATASET_ID are set

References

FLOW.md - Detailed operation flow
TESTCASE-FORMAT.md - Test case writing guide

---name: tc-reviewdescription: Review QA test case documents (Markdown) against source code. Use when: (1) Reviewing test cases for coverage gaps, (2) Learning source code to extract test scenarios, (3) Suggesting missing test cases, (4) Measuring test case quality. Supports multi-repo projects with dual memory (local MD + Dify RAG).---# Test Case Review AgentAI-powered test case document reviewer. Learns source code, then reviews test cases to find coverage gaps.## Quick Start```bash# Create project and add repotc-review project create myprojecttc-review project add-repo myproject https://github.com/org/repo.git# Learn source codetc-review learn myproject# Review test casestc-review review tests/checkout.md --project myproject```## Core Commands### Project Management```bashtc-review project create <name> # Create projecttc-review project list # List projectstc-review project show <name> # Show detailstc-review project add-repo <name> <url> # Add GitHub repotc-review project delete <name> # Delete project```### Learning & Review```bashtc-review learn <project> # Learn from all repostc-review review <file.md> --project <p> # Review test case filetc-review suggest <project> # Suggest new test cases```### Memory```bashtc-review memory status # Check local + Dify statustc-review memory search <proj> <query> # Search knowledgetc-review memory sync <project> # Sync to Dify RAG```### Config```bashtc-review config show # Show current configtc-review config init # Create config.jsontc-review config env # List env variables```## LLM Configuration### Anthropic Claude (Default)```bashexport LLM_PROVIDER=anthropicexport ANTHROPIC_API_KEY=sk-ant-...export LLM_MODEL=claude-sonnet-4-20250514```### OpenAI / Codex```bashexport LLM_PROVIDER=openaiexport OPENAI_API_KEY=sk-...export LLM_MODEL=gpt-4-turbo```## Test Case FormatTest cases must be in Markdown format:```markdown# Test Suite: Checkout Flow## TC-001: Successful Payment**Preconditions:**- User logged in- Cart has items**Steps:**1. Navigate to checkout2. Enter valid card3. Click Pay**Expected Result:**Payment success, order created```## Output Example```📊 Review Report: checkout.md━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Score: 65/100 Coverage: 65%✅ Covered: • Successful payment • Invalid card⚠️ Missing: • Payment timeout • Expired card • Concurrent payments💡 Suggested: TC-003: Payment Timeout TC-004: Expired Card Validation```## Workflows### Initial Setup1. Create project: `tc-review project create <name>`2. Add repos: `tc-review project add-repo <name> <url>`3. Learn: `tc-review learn <name>`4. (Optional) Sync to Dify: `tc-review memory sync <name>`### Review Cycle1. QA writes test cases in Markdown2. Run: `tc-review review <file> --project <name>`3. Address gaps found4. Re-review until score > 80### Get Suggestions```bashtc-review suggest myproject```Generates test case suggestions based on learned knowledge.## Memory System| Layer | Storage | Search | Best For ||-------|---------|--------|----------|| Local | MD + JSON | Keyword | Offline, fast || Remote | Dify RAG | Semantic | Large codebases |## Dify RAG Setup (Optional)```bashexport DIFY_ENABLED=trueexport DIFY_API_KEY=app-...export DIFY_DATASET_ID=...export DIFY_BASE_URL=https://api.dify.ai/v1```## Troubleshooting**No ANTHROPIC_API_KEY**: Set `export ANTHROPIC_API_KEY=sk-ant-...`**Project not found**: Run `tc-review project list` to see available projects**Low coverage score**: Check gaps in report, add missing test cases**Dify not connected**: Verify `DIFY_API_KEY` and `DIFY_DATASET_ID` are set## References- [FLOW.md](references/FLOW.md) - Detailed operation flow- [TESTCASE-FORMAT.md](references/TESTCASE-FORMAT.md) - Test case writing guide

Frequently asked questions