A setup workflow that adds Nightshift's overnight task loop to a code repository. It creates project files, safety checks, a stack-specific verifier, and a document for the work to be landed.

It helps scaffold an overnight plan, detect the project stack, reuse an existing plan when supplied, and verify whether the loop, checks, hooks, and landing files are already present.

It checks the repository before writing and leaves the automated loop switched off. This reduces the risk of adding duplicate setup or allowing unattended changes without the repository's own guards.

init — Skill for Claude Code by jasonm4130

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code

/plugin marketplace add jasonm4130/claude-skills

Claude Code

/plugin install nightshift

Made for: Claude Code.

Or install nightshift, the plugin that ships this one along with the rest of its 5 skills, 2 agents.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/jasonm4130/claude-skills/init/github.svg)](https://agentmods.dev/skills/jasonm4130/claude-skills/init)

Your own site

<a href="https://agentmods.dev/skills/jasonm4130/claude-skills/init"><img src="https://agentmods.dev/badge/skills/jasonm4130/claude-skills/init/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/jasonm4130/claude-skills/init"><img src="https://agentmods.dev/badge/skills/jasonm4130/claude-skills/init.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 180 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,215 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00180	$0.01215
Opus 5	$0.00090	$0.00607
Sonnet 5	$0.00036	$0.00243
Haiku 4.5	$0.00018	$0.00121

Measured 2d ago against content hash 151b66363094, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

init 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 2d 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.

plugins/nightshift/skills/init/SKILL.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Scaffold Nightshift into this repo

The loop is committed in the target repo, not in this plugin: it runs claude -p --setting-sources project, where installed plugins never load, so the hooks only bite when they live under the repo's own .claude/. This skill copies them in, proves the copy, and leaves the switch off. Announce: "Using nightshift:init to scaffold the overnight landing loop."

Scripts live under this skill's plugin directory: resolve ${CLAUDE_PLUGIN_ROOT} (or this file's ../../scripts/) and call them with node.

1. Look before writing

git remote get-url origin is a GitHub remote, gh auth status succeeds.
Detect the stack the way init.mjs --stack auto does (Cargo.toml, package.json, pyproject.toml, go.mod, else generic) and say which verifier skeleton it gets.
An existing loop/, scripts/check, or .claude/hooks/ means this is not a first init: run node <plugin>/scripts/init.mjs --check and stop with its report instead.
Is there a plan to land? If the user names one, pass --plan <path>; otherwise init scaffolds docs/plans/<today>-nightshift-smoke.md with one harmless task so the first night has something to do.

2. Scaffold

node <plugin>/scripts/init.mjs --stack <stack> [--plan <path>] [--base <branch>]

Say what it wrote (it prints the list): loop/*, .claude/hooks/*, scripts/check (never overwritten if present), docs/developing/landing.md or docs/nightshift.md, the smoke plan, and the merged .claude/settings.json. Then read loop/config back to the user: MERGE_MODE (protected when the base branch has required checks, else wait) and EXPECTED_CHECKS (gate only when a gate CI job exists; otherwise empty, and a human must fill it with the check names GitHub reports, which preflight lists). --deny-rules also adds permissions.deny entries; the hooks already deny in every permission mode.

3. Prove it, in this order

scripts/check from the repo root — last line must be CHECK OK. Edit the skeleton until it is (it is deliberately narrower than CI: fast, quiet).
loop/land.sh --dry-run — reads the plan from the checkout, so the uncommitted scaffold is enough. Expect STOP: frozen (the switch is unset) or, with LANDING_STATE=run already set, STOP: would run task 1: ….
node --test .claude/hooks/*.test.mjs — the copied guard tests pass in situ.
Commit the scaffold on a branch, one commit, and tell the user to open the PR: loop/, .claude/hooks/, .claude/settings.json, scripts/check, the docs page, the plan. Stage paths explicitly. The user opens the PR; this skill never pushes to the base branch.
node <plugin>/scripts/preflight.mjs — one line per check. Its plan check is expected to FAIL at this point (the plan is on the scaffold branch, not on origin/<base>) and says "merge the PR that carries it". Everything else should be ok or warn; a FAIL on protection or checks is a loop/config edit to make now.

Read the full file on GitHub · 82 lines

Changes

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.

2d ago Changed · +16 tokens per session 151b66363094
3d ago First seen · 82 lines · 164 tokens per session scan A fb93df91fa47

Subscribe to this mod's changes

init is a skill published in the GitHub repository jasonm4130/claude-skills (5 stars, last pushed 2d ago), licensed MIT. It adds 180 tokens to every session and 1,215 once invoked, about $0.0009 per session on Opus 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 2026-09-05.

Claude Code Project management ai-agents claude-code-plugin claude-skills plugin-marketplace

catchup

Summarize and review what changed while you were away. Use after a weekend, vacation, or flight to check missed PRs, git commits, Linear tickets, and meetings — one prioritized brief, not a firehose.

oliver-kriska/claude-elixir-phoenix · 48 tokens

plan

Plan features spanning multiple domains: billing (Stripe), auth (RBAC), real-time (Presence), webhooks, jobs (Oban). Use when designing interconnected systems or converting review findings into tasks.

oliver-kriska/claude-elixir-phoenix · 42 tokens

work

Execute Elixir/Phoenix plan tasks with progress tracking. Use after /phx:plan to implement features with mix compile and mix test verification after each step, or --continue to resume interrupted work.

oliver-kriska/claude-elixir-phoenix · 43 tokens

phx-deps-update

Bump outdated Hex deps — inventory, snapshot changelogs, update, fix breaks, split reviewable PRs (patches bundled, majors solo). Use to upgrade/bump Elixir dependencies or when versions fall behind. NOT for deps.get failures (phx-investigate).

oliver-kriska/claude-elixir-phoenix · 62 tokens

customer-success-management

Runs the ongoing relationship with accounts after the sale — segmenting coverage against account value, building a health score that predicts rather than describes, running reviews customers find worth attending, forecasting renewals honestly, and finding expansion that follows usage instead of quota. Use this to…

cbrock84/headcount · 85 tokens

chief-information-officer

The CIO's remit — running the technology the company works on, service quality, IT spend, and the boundary with product engineering. Use this to set IT priorities, decide what IT owns versus engineering, structure IT spend or an IT roadmap, judge whether to build, buy or outsource, or work out why IT is seen as a cost…

cbrock84/headcount · 79 tokens