Automated summary
Captures the output of a tmux pane running tests so an agent can drive a TDD loop (RED/GREEN/REFACTOR) in any language.
Strengths
- Reads test output straight from the pane with `tmux capture-pane`, no need to re-run the suite
- Detection patterns provided for Rust, Python, JavaScript and Go (failures, assertions, compile errors)
- Ready-made commands to send a command, interrupt a process or list panes
- Explicit TDD loop that checks test status both before and after each change
Limitations
- Assumes a fixed tmux layout (pane 0.1 is the runner) and an already-running watcher
- Failure analysis is purely textual: no structured parsing or automatic mapping to source files
- Does not cover installing or setting up tmux, the watcher, or the initial session
Use it when a test suite is already running in a tmux pane and the agent needs to read its output to iterate in TDD fashion.
Avoid it when the agent runs tests in the foreground itself, or when no tmux environment is available.
Security analysis
CautionThe skill only instructs reading test output (tmux capture-pane) and sending keystrokes (tmux send-keys) to a local test pane. There is no network exfiltration, credential access, destructive deletion, or encoded payload. The main residual risk is that send-keys can execute any shell command in a pane and the skill declares no tool restrictions, so it is powerful-but-legitimate rather than safe.
- •Uses tmux send-keys to inject arbitrary commands into another pane (legitimate but powerful; could run unintended shell commands if pane content is misinterpreted)
- •Relies on shell/terminal access with no declared allowed-tools, leaving execution scope unbounded
- •No destructive, exfiltration, or obfuscation patterns present
Examples
Capture pane 0.1 in tmux and tell me whether the test suite is currently passing or failing, and list the failing tests.The test for parse_duration is failing in the tmux test pane. Read the output, write the minimal code to make it pass, then re-capture the pane to confirm it's green.Send `go test ./...` to the test pane in tmux, then read the last 50 lines and summarize which packages failed.name: tmux-tdd description: Interact with tmux-based TDD environments. Use when running tests in tmux panes, checking test output, parsing test failures, or orchestrating TDD workflows. Works with any language (Rust, Python, JavaScript, Go, etc). Read test output directly from tmux panes.
tmux TDD Skill
Orchestrate Test-Driven Development in a tmux environment by reading test output directly from panes. Language agnostic.
Environment Layout
┌────────────────┬────────────────┐
│ │ │
│ Pane 0.0 │ Pane 0.1 │
│ Claude Code │ Test Runner │
│ │ │
└────────────────┴────────────────┘
- Pane 0.0: Claude Code (where you are)
- Pane 0.1: Test runner (cargo watch, pytest, jest, go test, etc.)
Instructions
Step 1: Check Current Test Status
Capture output directly from the test pane:
tmux capture-pane -t 0.1 -p -S -50
Options:
-t 0.1: Target pane (test runner)-p: Print to stdout-S -50: Last 50 lines (use-S -for full history)
Step 2: Parse Test Results
Look for pass/fail patterns based on the language:
Rust (cargo test):
test result: ok. 6 passed; 0 failed
test result: FAILED. 4 passed; 2 failed
error[E0425]: cannot find value `x` in this scope
Python (pytest):
6 passed in 0.12s
2 failed, 4 passed in 0.15s
FAILED tests/test_app.py::test_add - AssertionError
E assert 4 == 5
JavaScript (jest/vitest):
Tests: 6 passed, 6 total
Tests: 2 failed, 4 passed, 6 total
FAIL src/app.test.js
✕ should add numbers (5 ms)
Go (go test):
ok mypackage 0.005s
FAIL mypackage 0.005s
--- FAIL: TestAdd (0.00s)
app_test.go:10: expected 5, got 4
Step 3: Make Code Changes
Edit the source files to fix failures. If using a watch mode, tests will automatically re-run.
Step 4: Verify Fix
Wait for tests to re-run, then capture pane again:
tmux capture-pane -t 0.1 -p -S -30
Step 5: Send Commands to Test Pane (if needed)
# Run command in test pane
tmux send-keys -t 0.1 "your-test-command" C-m
# Stop running process
tmux send-keys -t 0.1 C-c
Common Test Commands by Language
| Language | Watch Mode | Single Run |
|----------|------------|------------|
| Rust | cargo watch -x test | cargo test |
| Python | ptw or pytest-watch | pytest |
| JavaScript | jest --watch or vitest | jest or npm test |
| Go | gotestsum --watch | go test ./... |
| Ruby | guard | rspec |
Quick Reference
| Task | Command |
|------|---------|
| Read test output | tmux capture-pane -t 0.1 -p -S -50 |
| Full history | tmux capture-pane -t 0.1 -p -S - |
| Send command | tmux send-keys -t 0.1 "command" C-m |
| Stop process | tmux send-keys -t 0.1 C-c |
| List panes | tmux list-panes |
TDD Workflow
- RED: Write a failing test, verify it fails via pane capture
- GREEN: Write minimal code to make it pass
- REFACTOR: Clean up while keeping tests green
Always check test status before and after making changes.
Troubleshooting
Empty pane output?
- Test runner may not be running
- Check:
tmux capture-pane -t 0.1 -p -S -5 - Restart the test watcher in pane 0.1
Can't connect to pane?
- Verify tmux session exists:
tmux list-sessions - List panes:
tmux list-panes
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/tmux-tdd-orchestration-mmbizkvh' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/tmux-tdd-orchestration-mmbizkvh/download' --output '.skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md.part' && mv '.skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md.part' '.skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
Original excerpt: Quick Reference
| Task | Command |
|---|---|
| Read test output | tmux capture-pane -t 0.1 -p -S -50 |
| Full history | tmux capture-pane -t 0.1 -p -S - |
| Send command | tmux send-keys -t 0.1 "command" C-m |
| Stop process | tmux send-keys -t 0.1 C-c |
| List panes | tmux list-panes |
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md' && test -s '.skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/tmux-tdd' && cp -R '.skill-downloads/tmux-tdd-orchestration-mmbizkvh' '.claude/skills/tmux-tdd'Codex
test -f '.skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md' && test -s '.skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md' && mkdir -p '.agents/skills' && test ! -e '.agents/skills/tmux-tdd' && cp -R '.skill-downloads/tmux-tdd-orchestration-mmbizkvh' '.agents/skills/tmux-tdd'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/tmux-tdd-orchestration-mmbizkvh'openskills read 'tmux-tdd-orchestration-mmbizkvh'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/tmux-tdd-orchestration-mmbizkvh/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
ClickHouse Performance Comparison
Testing
Evaluate ClickHouse performance test results, check PR performance changes, compare with history, and assess whether a result is actionable.
Analyze CI Performance Reports
Testing
Analyze CI performance comparison reports for a ClickHouse PR. Cross-references master history to distinguish real regressions from flaky tests.
Santa Method
Testing
Multi-agent adversarial verification with two independent reviewers. Output is checked twice and iteratively fixed until both pass before shipping.