speckit-scripttest

Generate Bash test scripts from spec-kit test plan markdown files using parallel sub-agents.

Updated Jul 13, 2026
One-click install
npx skills add https://github.com/torikhayes/aeai-dotnet-brownfield --skill speckit-scripttest-torikhayes
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: speckit-scripttest
Source: https://github.com/torikhayes/aeai-dotnet-brownfield/tree/main/.claude/skills/speckit-scripttest
Command: npx skills add https://github.com/torikhayes/aeai-dotnet-brownfield --skill speckit-scripttest-torikhayes

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manually converting test plan markdown files into executable shell scripts is slow and error-prone. This Skill automates the translation of TC-### test cases into runnable Bash scripts with prerequisite checks and assertions. ## Core Features & Use Cases - Automated Script Generation: Reads test-plan.md and checklist files, then generates one tc-{NNN}.sh Bash script per test case with setup, assertions, and PASS/FAIL reporting. - Parallel Sub-Agent Processing: Spawns one sub-agent per markdown source file so large test plans are converted without blocking the main conversation. - Orchestrator Generation: Produces a run-all.sh script that executes all test cases, supports filtering by name, and prints a pass/fail/skip summary. - Use Case: After running /speckit-testplan to produce a test plan with 30 test cases, invoke this Skill to generate executable integration test scripts that verify API responses with curl, database state with psql, and JSON fields with jq. ## Quick Start Ask the AI to generate executable test scripts from the current feature's test plan, optionally filtering to a specific file or TC number.

Frequently Asked Questions about speckit-scripttest

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert a test plan markdown into executable shell scripts?▼

Run this Skill after generating test-plan.md with /speckit-testplan. It reads every TC-### heading, translates steps into Bash commands with curl, psql, and jq assertions, and writes one tc-{NNN}.sh script per test case plus a run-all.sh orchestrator.

How to run only specific test cases from a generated test suite?▼

Pass a file name or TC filter as the argument, such as 'api' or 'TC-010', to process only matching markdown files or test cases. The generated run-all.sh also accepts script names like './run-all.sh tc-010 tc-011' to run a subset.

Does this work without a spec-kit project structure?▼

No. It requires a spec-kit project with a .specify/ directory and an existing test-plan.md containing TC-### headings. If no test cases are found, it stops and instructs you to run /speckit-testplan first.

What happens when a test case cannot be fully automated?▼

Test cases with purely manual or visual steps get a MANUAL comment in the generated script instead of assertions. The final summary report lists all partially automated TCs so you know which ones still need human verification.

Why does a generated test script skip instead of failing?▼

Scripts verify prerequisite data before the test body, such as auth tokens, seeded rows, or fixture data. If a prerequisite like $TOKEN is missing, the script echoes SKIP and exits cleanly rather than producing a misleading failure.