Productivity
Skill
/red-tests
Feeds every guard in Scripts/ something it must reject and confirms it says no, then reports the count exactly as printed with the skips named; before a release the same suite runs against the staged tree and blocks the build while any guard has not gone red. Use when the user
Claude Code
$ npx -y skills add myICOR/myPKA --skill red-tests --agent claude-code
Codex
$ npx skills add myICOR/myPKA --skill red-tests -a codex -y
Cursor
$ npx skills add myICOR/myPKA --skill red-tests -a cursor -y
Gemini CLI
$ npx skills add myICOR/myPKA --skill red-tests -a gemini-cli -y
GitHub Copilot
$ npx skills add myICOR/myPKA --skill red-tests -a github-copilot -y
OpenCode
$ npx skills add myICOR/myPKA --skill red-tests -a opencode -y
Windsurf
$ npx skills add myICOR/myPKA --skill red-tests -a windsurf -y
SKILL.md
red-tests.SKILL.mdname: red-tests
description: "Feeds every guard in Scripts/ something it must reject and confirms it says no, then reports the count exactly as printed with the skips named; before a release the same suite runs against the staged tree and blocks the build while any guard has not gone red. Use when the user says: \"run the red tests\", \"red-test the guards\", \"did every guard go red\", \"is the release gate green\", \"check the guards before the release\"."
disable-model-invocation: true
user-invocable: true
shell: bash
allowed-tools: Bash
<!-- GENERATED by scaffold-init.py from `06 AI Team/AI Team Knowledge/SOPs/SOP-1016-run-the-red-tests-and-gate-a-release.md`. content-hash:7cf0ec79aa7c. Do not hand-edit: change the source and re-run the generator. -->
You are Mack. If you are the orchestrator, hand this to the specialist `mack`; if you are already Mack, act directly.
Run this only when the user asks for it by name.
Read `06 AI Team/AI Team Knowledge/SOPs/SOP-1016-run-the-red-tests-and-gate-a-release.md` now and follow it exactly. That file is the procedure; this file is the door.
Run `python3 "06 AI Team/AI Team Knowledge/Scripts/run-red-tests.py"` with your shell tool, from the folder that holds `AGENTS.md`, before step 1, unless its output is already below. Read that output; do not run it twice.
!`python3 "${CLAUDE_PROJECT_DIR}/06 AI Team/AI Team Knowledge/Scripts/run-red-tests.py"`
Script calls the procedure names, by path:
- `06 AI Team/AI Team Knowledge/Scripts/run-red-tests.py`
- `06 AI Team/AI Team Knowledge/Scripts/release-gate-red-tests.sh`
- `06 AI Team/AI Team Knowledge/Scripts/build-mypka-release.sh`
- `06 AI Team/AI Team Knowledge/Scripts/build-release-zip.sh`
Resources (open only when a step names them):
- `06 AI Team/Agents/Mack/AGENT.md`
- `06 AI Team/AI Team Knowledge/SOPs/SOP-1016-run-the-red-tests-and-gate-a-release.md`
Return the completion evidence the procedure's last step names, and nothing it does not.
Read more
name: red-tests description: "Feeds every guard in Scripts/ something it must reject and confirms it says no, then reports the count exactly as printed with the skips named; before a release the same suite runs against the staged tree and blocks the build while any guard has not gone red. Use when the user says: \"run the red tests\", \"red-test the guards\", \"did every guard go red\", \"is the release gate green\", \"check the guards before the release\"." disable-model-invocation: true user-invocable: true shell: bash allowed-tools: Bash
<!-- GENERATED by scaffold-init.py from `06 AI Team/AI Team Knowledge/SOPs/SOP-1016-run-the-red-tests-and-gate-a-release.md`. content-hash:7cf0ec79aa7c. Do not hand-edit: change the source and re-run the generator. -->
You are Mack. If you are the orchestrator, hand this to the specialist `mack`; if you are already Mack, act directly.
Run this only when the user asks for it by name.
Read `06 AI Team/AI Team Knowledge/SOPs/SOP-1016-run-the-red-tests-and-gate-a-release.md` now and follow it exactly. That file is the procedure; this file is the door.
Run `python3 "06 AI Team/AI Team Knowledge/Scripts/run-red-tests.py"` with your shell tool, from the folder that holds `AGENTS.md`, before step 1, unless its output is already below. Read that output; do not run it twice.
!`python3 "${CLAUDE_PROJECT_DIR}/06 AI Team/AI Team Knowledge/Scripts/run-red-tests.py"`
Script calls the procedure names, by path:
- `06 AI Team/AI Team Knowledge/Scripts/run-red-tests.py`
- `06 AI Team/AI Team Knowledge/Scripts/release-gate-red-tests.sh`
- `06 AI Team/AI Team Knowledge/Scripts/build-mypka-release.sh`
- `06 AI Team/AI Team Knowledge/Scripts/build-release-zip.sh`
Resources (open only when a step names them):
- `06 AI Team/Agents/Mack/AGENT.md`
- `06 AI Team/AI Team Knowledge/SOPs/SOP-1016-run-the-red-tests-and-gate-a-release.md`
Return the completion evidence the procedure's last step names, and nothing it does not.
Ships withmypka
Your life and work in one folder of plain markdown, organised by ICOR, with an AI team that works in it with you. Open it in Claude Code, Codex, Gemini CLI or Cursor, or in Obsidian, or just a text editor. myPKA 7: one folder again.
3199 skills14 agents1 MCP
Stars on GitHub
Momentum
- stars in total
- 319
Other skills on mypka.
More skills from other plugins.
i-have-adhd
Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress…
@ayghriInstall Skill
distilly
Set up the owner profile so a personal agent knows who it works for (first-run setup, \"get to know me\"), or distill colleague,…
@titanwingsInstall SkillQuestions about red-tests.
- Which coding agents can use red-tests?
- red-tests installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.