StarNet is a local-first desktop harness for creating AI agents and arranging them in a pixel-art space station whose layout represents their capabilities, permissions, and workflow. It lets people run real model calls and tools through multiple bounded agents, using their own provider keys or accounts and optionally messaging the station through chat services. The catalogue entries are skills that extend or support work performed by StarNet agents.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add androoAGI/starnet --skill finishing-a-branchgit clone --depth 1 https://github.com/androoAGI/starnetWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/androoagi/starnet/finishing-a-branch)<a href="https://agentmods.dev/skills/androoagi/starnet/finishing-a-branch"><img src="https://agentmods.dev/badge/skills/androoagi/starnet/finishing-a-branch/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/androoagi/starnet/finishing-a-branch"><img src="https://agentmods.dev/badge/skills/androoagi/starnet/finishing-a-branch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00052 | $0.02437 |
| Opus 5.5 | $0.00021 | $0.00975 |
| Sonnet 5.5 | $0.00010 | $0.00487 |
| Haiku 4.5 | $0.00005 | $0.00244 |
Grade A, and why
finishing-a-branch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Finishing a Branch
Overview
Core principle: verify tests → detect the environment → present the options → carry out the choice → clean up.
Use this when the implementation on a branch is complete and the question is how to integrate it. The integration decision belongs to the Commander; your job is to make it safe and easy.
All commands run through shell.exec (tests through verify.run). On Windows shell.exec runs cmd.exe, so do not rely on shell variables: run each git command, read its output, and carry the values yourself. To run from another folder, pass that folder as cwd; do not cd with absolute paths.
If the project has its own written merge rules (a contributing guide, a repo protocol file), those rules win over this menu. Read them first.
Step 1: Verify tests
Run the project's full test suite with verify.run (cmd: npm test, cargo test, pytest, go test ./..., or whatever the project uses). Read the whole result.
If tests fail, report the failures and stop. The menu comes only after a green suite:
Tests failing (<N> failures). Must fix before completing:
[the failures]
If tests pass: continue to Step 2.
Step 2: Detect the environment
Run these and write down the answers now, while you are still inside the workspace (Step 5 moves to another folder, and Step 6 needs these values):
| Command | You learn |
|---|---|
git rev-parse --absolute-git-dir |
GIT_DIR |
git rev-parse --path-format=absolute --git-common-dir |
GIT_COMMON |
git rev-parse --show-toplevel |
WORKTREE_PATH (this workspace) |
git branch --show-current |
the branch name; empty means a detached HEAD |
git worktree list |
the first line is the main repo root (MAIN_ROOT) |
git status --porcelain |
uncommitted work; commit it or ask before going on |
This decides which menu to show and how cleanup works:
| State | Menu | Cleanup |
|---|---|---|
| GIT_DIR equals GIT_COMMON (normal repo) | Standard 3 options | No worktree to clean up |
| GIT_DIR differs, named branch | Standard 3 options | Depends on who made the worktree (Step 6) |
| GIT_DIR differs, detached HEAD | Reduced 2 options (no merge) | Managed from outside: leave it in place |
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 220 lines · 52 tokens per session scan A 6a1cfa653bab
finishing-a-branch is a skill published in the GitHub repository androoAGI/starnet (1,468 stars, last pushed 3d ago), licensed MIT. It adds 52 tokens to every session and 2,437 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-04.
Other skills, from other repositories
PR babysit (watch & drive to green)
Watch a pull request and delegate fixes until CI is green and reviews are addressed. Requires the GitHub MCP server or gh.
precommit
Pre-commit checks — lint:fix -> build -> test.
precommit-fast
Quick pre-commit checks — lint:fix -> test.
Check work (verify against criteria)
Verify an implementation against acceptance criteria with a reviewer and a tester.
Commit (clean, conventional)
Stage the right changes and write a clear, conventional commit message.
Write tests
Add focused, meaningful tests for code — happy path plus the edge cases that matter.