run-full-suite

Discovers and runs all speckit shell test suites across feature specs, then prints a consolidated cross-spec summary.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Running integration tests across many feature specs one by one is tedious and makes it hard to see overall project health. This Skill discovers every speckit shell test suite in the repository, executes them in sequence, and produces a single consolidated pass/fail report across all specs. ## Core Features & Use Cases - Automatic Suite Discovery: Scans specs/*/spec-integration-tests/ for run-all.sh and tc-*.sh files, counting test cases and checking for config.env. - Filtering and Preview Modes: Run only suites matching a spec name or number, preview suites with --list, or halt early with --fail-fast. - Environment Refresh: Runs detect-ports.sh --update before each suite so config.env reflects the current Aspire AppHost ports and connection strings. - Consolidated Reporting: Parses each suite's TEST SUMMARY block and prints a cross-spec table with totals, failure details, and exit guidance. - Use Case: After restarting the .NET Aspire AppHost, run the full suite to verify every feature spec still passes against the new ports and connection strings, then isolate failures per spec. ## Quick Start Ask the assistant to run the full test suite across all specs, optionally filtering by a spec number like 004 or using --list to preview discovered suites.

Frequently Asked Questions about run-full-suite

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

FAQPage Schema
How do I run all integration test suites across multiple feature specs at once?▼

Invoke the run-full-suite skill with no arguments to discover every run-all.sh under specs/ and execute each suite in sequence. It prints a consolidated cross-spec summary with pass, fail, skip, and total counts.

How can I run tests for only one specific spec?▼

Pass a filter argument such as a spec number or name fragment, for example 004 or token. Only suites whose spec folder name contains the filter string (case-insensitive) will run.

Can I preview discovered test suites without running them?▼

Yes, pass the --list flag to print a discovery table showing each spec, its tc-*.sh count, whether config.env exists, and whether the suite is runnable, without executing anything.

Why do tests fail after restarting the Aspire AppHost?▼

Aspire reassigns ports and passwords on every restart, so a stale config.env causes spurious failures. The skill runs detect-ports.sh --update before each suite to refresh connection strings, but the AppHost must be running for detection to succeed.

What tools are required for the shell test suites to run?▼

The suites rely on curl for HTTP tests, jq for JSON assertions, psql for database tests, and docker for port detection. Missing tools produce warnings but do not block execution; affected tests report their own failures.

How do I stop the full suite run at the first failure?▼

Pass the --fail-fast flag. Execution halts after the first suite with any failed test case, prints that suite's failure details, and shows a partial cross-spec summary noting the early stop.