yolo-jail: Skill for Claude Code

.pi/skills/cutting-releases/SKILL.md

cutting-releases is a skill for Claude Code, Codex from mschulkind-oss/yolo-jail. It costs 34 tokens per session (3,399 once invoked), scanned A, original, Apache-2.0.

A release-preparation workflow for the yolo-jail project. It covers release notes, version preparation, release checks, and verifying publication, while treating publication as a separate action requiring explicit permission.

In plain words
What is it for?
It is for preparing release versions, writing release notes, checking release gates, handling failed or partial releases, and verifying published artifacts.
Why use it?
It reduces mistakes around tags, changelogs, repeated test runs, and accidental publication. It also makes clear which checks have actually run and which are still pending.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions AGENTS.md.

This is mschulkind-oss/yolo-jail's own configuration. It tells Claude Code and Codex how to work on yolo-jail itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything yolo-jail configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is version=$(sh scripts/release-version.sh).

Reuse

Borrowing it

Nothing to install: this file belongs to mschulkind-oss/yolo-jail. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/mschulkind-oss/yolo-jail/main/.pi/skills/cutting-releases/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/mschulkind-oss/yolo-jail

Made for: Claude Code, Codex.

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.

agentmods badge for cutting-releases

README.md
[![agentmods](https://agentmods.dev/badge/skills/mschulkind-oss/yolo-jail/cutting-releases/github.svg)](https://agentmods.dev/skills/mschulkind-oss/yolo-jail/cutting-releases)
Your own site
<a href="https://agentmods.dev/skills/mschulkind-oss/yolo-jail/cutting-releases"><img src="https://agentmods.dev/badge/skills/mschulkind-oss/yolo-jail/cutting-releases/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.

agentmods 80×15 button for cutting-releases

Your own site · 80×15
<a href="https://agentmods.dev/skills/mschulkind-oss/yolo-jail/cutting-releases"><img src="https://agentmods.dev/badge/skills/mschulkind-oss/yolo-jail/cutting-releases.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,399 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00034 $0.03399
Opus 5.5 $0.00014 $0.01360
Sonnet 5.5 $0.00007 $0.00680
Haiku 4.5 $0.00003 $0.00340

Measured 3d ago against content hash cce247817de1, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

cutting-releases 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 3d 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.

.pi/skills/cutting-releases/SKILL.md · 96 lines

How it starts

The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cut a yolo-jail release

Use this workflow for yolo-jail releases. Read the release-process reference when a request meets an existing tag, failed workflow, partial publication, or resume. Write the changelog section by the rules below. Landing work adds no changelog entry, so the section is written here, once, from the commits since the last tag.

A release request can create an immutable tag and publish artifacts. Never run just release or dispatch a workflow unless the owner explicitly authorized publication. This skill prepares and verifies; it does not grant publication authority.

Reuse evidence; do not repeat the full suite

Keep code landing, release preparation, and publication as separate stages:

  • Code changes follow the normal testing and landing rules in AGENTS.md. Normally, a launch-path landing includes one full integration suite. If the owner explicitly orders a particular code landing main-first, do not make a duplicate local suite a precondition to that already-authorized landing; report local and CI checks honestly as pending or canceled. This is a narrow sequencing instruction, not a general waiver of checks or hooks. The exact target SHA's regular main-push CI, including integration proof, is still required before tagging—but the release request runs that gate automatically; it is not a manual precondition to invoking just release.
  • Regular ci.yml push CI runs the complete integration suite across Linux architectures and shards, then checks that every listed test ran. Before tagging, the release gate requires a successful regular main push CI run for the exact target SHA, including all required jobs and shard-completeness proof. A green PR run alone is not enough. The gate is enforced by preflight.sh and gate.go. preflight.sh invokes the gate with a 60-minute timeout and 15-second polling interval, first in release-request eligibility and again immediately before the tag write. The gate waits while proof is pending or absent; a completed failed/cancelled run or a proof/API timeout refuses without creating the tag. Release workflows consume that exact-SHA proof; release and publish recheck it. They prepare and publish release outputs but do not rerun the integration suite. If only release notes, docs, or the version heading changed, do not add another full Go or integration run: validate the notes locally, then rely on the exact-SHA main-push CI gate. If code, tests, or release-gate inputs changed, run the relevant tests under AGENTS.md and require CI for the final target SHA.

Read the full file on GitHub · 96 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.

  1. 3d ago Changed cce247817de1
  2. 4d ago Changed · +16 lines 33d6305bdb2f
  3. 7d ago First seen · 80 lines · 34 tokens per session scan A 48fe823583ca

Subscribe to this mod's changes

cutting-releases is a skill published in the GitHub repository mschulkind-oss/yolo-jail (3 stars, last pushed today), licensed Apache-2.0. It adds 34 tokens to every session and 3,399 once invoked, about $0.0001 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 2026-10-21.