Squid Implement Night logo

Squid Implement Night

Community
iusztinpaul
squid-implement-night

Run the full agent-team pipeline end-to-end for one feature whose Tasks Plan is already approved by /squid-plan, handing the human a validated, ready-to-squash-merge PR. Trigger after /squid-plan.

Overview

Publisheriusztinpaul
Repositorysquid
Skill namesquid-implement-night
Stars
207
Forks
34
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by iusztinpaul on GitHub. Read the source before you install it.

Installation

Install the Squid Implement Night AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/iusztinpaul/squid/tree/main/skills/squid-implement-night
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/iusztinpaul/squid.git /tmp/squid
mkdir -p .claude/skills
cp -r /tmp/squid/skills/squid-implement-night .claude/skills/squid-implement-night
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Squid Implement Night in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Squid Implement Night on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Squid Implement Night is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Implement Night — end-to-end feature pipeline (thin orchestrator)

Build an already-approved feature plan all the way to a validated, ready-to-merge PR. This skill is a thin orchestrator: the real work lives in the sub-skills it invokes — /squid-implement-task, /squid-review, /squid-review-ci — plus the agents those skills launch.

$ARGUMENTS is the approved Tasks Plan reference (path or feature slug) produced by /squid-plan. If empty, ask the human which plan to build (and remind them to run /squid-plan first if none exists).

You are the orchestrator — a MANAGER. You sequence the sub-skills, route rollback tasks, and enforce the gates. You do NOT write code, review the diff, or run tests yourself.

Read AGENTS.md first (tracker mode, the pipeline map, and the cross-cutting rules). Retry caps live in each sub-skill.

Input: an approved Tasks Plan + the feature branch/worktree /squid-plan created. Output: a CI-validated feature PR, ready for the human to squash-merge.

The pipeline blocks on the human only at the very end (squash-merge). (Plan approval already happened in /squid-plan.) Everything between is automated; failures route back into /squid-implement-task as rollup tasks rather than stopping the pipeline. Caps stop the pipeline: Tester FAIL 5/task, PA REJECT 3, PR-Reviewer 3, On-Call 5 — when a cap fires, surface USER ACTION REQUIRED and stop.

Critical rules:

  • Never rubber-stamp a sub-skill's result. When /squid-review reports "no blockers" or /squid-review-ci reports "green", confirm the evidence is real before advancing.
  • By name, never by path. Invoke /squid-implement-task, /squid-review, /squid-review-ci by their skill names so it works when installed as a plugin.

Step 0 — Locate the branch worktree + tasks

/squid-plan created the feature branch feat/{slug} — either in a dedicated worktree at ../{repo}-{slug} or in the current working tree — and wrote the feature's task files (tasks/<NNN>-*.md, status: pending) on it. Find where that branch is checked out:

bash
git worktree list          # find the working tree whose branch is feat/{slug}

Confirm that working tree exists and contains pending task files (tasks/<NNN>-*.md with status: pending); WORKTREE_PATH is its path (the dedicated worktree, or the repo root if the branch lives in the current tree). If there's no matching branch or no pending tasks, stop and tell the human to run /squid-plan first. Pass Working directory: {WORKTREE_PATH} to every sub-skill / agent so all work happens there.


Step 1 — Build the whole plan

Invoke /squid-implement-task once with the feature's pending tasks. It loops over every task — SWE → Tester → commit on PASS → next. (Do NOT loop here yourself; the per-task iteration lives inside /squid-implement-task.)

invoke /squid-implement-task with: the feature's pending tasks (tasks/<NNN>-*.md, status: pending), Working directory: {WORKTREE_PATH}

Step 2 — Review (loop until clean)

invoke /squid-review with: feature {title}, Working directory: {WORKTREE_PATH}

/squid-review pushes, creates/updates the PR, and runs PA acceptance then PR-Reviewer.

  • Returns "clean" (no blockers) → proceed to Step 3.
  • Returns a rollup task (PA REJECT or PR-Reviewer Blockers) → invoke /squid-implement-task on that one rollup task (it builds + commits the fix), then re-invoke /squid-review. Repeat.

Step 3 — Review CI

invoke /squid-review-ci with: PR #{N}, Working directory: {WORKTREE_PATH}

/squid-review-ci watches CI and drives it green (On-Call diagnoses → SWE fixes → re-check).

  • CI green → the PR is validated and ready to merge. Proceed to the tail.

Tail — hand-off

The validated PR is ready. The orchestrator does NOT squash — per-task commits stay on the branch; the human uses GitHub's "Squash and merge".

Report the final summary:

markdown
## /squid-implement-night complete — {Feature title}

**PR:** {URL} (validated; ready to squash-merge)
**Branch:** feat/{slug} ({N} per-task commits)
**Worktree:** {WORKTREE_PATH}

**Tasks delivered ({N}):** {table — Tester / PA accept / PR-Reviewer / CI}
**Rollup tasks ({M}):** {list, or "none"}

Next: review the PR, then GitHub's **Squash and merge**. Remove the worktree with `git worktree remove {WORKTREE_PATH}` after merge.

/squid-implement-night ends here.


Notes on shape

  • /squid-review-ci handles its own CI fixes internally and does not re-enter /squid-review.
  • Don't add mid-pipeline gates.

Frequently asked questions

What does the Squid Implement Night AI skill do?

Run the full agent-team pipeline end-to-end for one feature whose Tasks Plan is already approved by /squid-plan, handing the human a validated, ready-to-squash-merge PR. Trigger after /squid-plan.

Why use Squid Implement Night on TypingMind?

Because you install it once and use it with any model. Squid Implement Night is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Squid Implement Night in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/iusztinpaul/squid/tree/main/skills/squid-implement-night. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Squid Implement Night?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Squid Implement Night?

As many as you like. As long as a model supports skills, you can use Squid Implement Night with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Squid Implement Night AI skill free?

Yes. It is published on GitHub by iusztinpaul under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇