Borrowing it
Nothing to install: this file belongs to kaltinril/ShadowDusk. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/kaltinril/ShadowDusk/main/.claude/skills/release/SKILL.mdgit clone --depth 1 https://github.com/kaltinril/ShadowDuskWrote 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/kaltinril/shadowdusk/release)<a href="https://agentmods.dev/skills/kaltinril/shadowdusk/release"><img src="https://agentmods.dev/badge/skills/kaltinril/shadowdusk/release.svg" alt="Measured on agentmods" 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 | $0.00073 | $0.03288 |
| Opus 5 | $0.00036 | $0.01644 |
| Sonnet 5 | $0.00015 | $0.00658 |
| Haiku 4.5 | $0.00007 | $0.00329 |
Grade A, and why
release 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 yesterday.
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 — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Release
Automate the full ShadowDusk release from version bump through PR merge to publish trigger.
A release publishes all eight ShadowDusk.* packages (Core, HLSL, GLSL, ShaderToy,
Compiler, Cli, Wasm, MgcbPlugin) plus the ShadowDuskCLI dotnet tool to nuget.org, and attaches
self-contained CLI binaries to a GitHub Release. The human runbook this automates is RELEASING.md.
Input
$ARGUMENTS is the version (e.g., 0.2.0). If omitted, read the current <Version> from
Directory.Build.props and ask the user for the new one.
Choosing the version bump (0.y.z, pre-1.0 — verified against the actual CHANGELOG.md history, not just SemVer's letter):
- MINOR (
0.X.0) — the release ships a genuine new capability: a new backend/target, a new frontend, a new consumer-facing feature. - PATCH (
0.X.Y) — fixes/maintenance only: bug fixes, doc corrections, CI/tooling reliability. A patch release can still carry a small accompanying addition (a README, a regression fixture) as long as the headline is a fix, not a feature. - MAJOR (1.0.0) is a deliberate decision the owner makes, not a numbering consequence of
accumulated features — see
plan/plan.md's named v1.0-gate phases for what "ready" means. - If a release doesn't cleanly fit one bucket, ask the user rather than guessing.
Steps
-
Validate clean tree.
git status; if dirty, warn and stop. Show current<Version>fromDirectory.Build.props. Thengit checkout main && git pullso the gates in step 2 run against exactly the code that is about to ship. -
RENDER GATES FIRST — DX / FNA / KNI / Vulkan (required; CI cannot do this). Run BEFORE the version bump: it is the longest and likeliest step to fail, it is the only proof of the actual product promise, and
release.ymldoes not check it, so a divergence must stop the release before any version churn, commit, PR, or CI time is spent.Run these from PowerShell 7 (
pwsh), not Windows PowerShell 5.1. The script shells out totools/restore.ps1, which carries#requires -Version 7.0; under 5.1 the run dies at the restore step with aScriptRequiresUnmatchedPSVersionerror that looks nothing like a render divergence.
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.
- yesterday First seen · 176 lines · 73 tokens per session scan A 1107803a1233
release is a skill published in the GitHub repository kaltinril/ShadowDusk (14 stars, last pushed 23d ago), licensed MIT. It adds 73 tokens to every session and 3,288 once invoked, about $0.0004 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 2026-09-04.
Other skills, from other repositories
agent-release-swarm
Agent skill for release-swarm - invoke with $agent-release-swarm.
release-announcement
Write a release announcement — changelog, blog post, in-app note, or social post — that leads with user impact, names the audience, and includes upgrade/migration steps without filler.
multi-agent-release-manager
Cleans up the workspace, formats code, runs presubmit checks, and uploads CLs to Gerrit.
release-notes
Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release notes, creating changelogs, announcing product updates, or summarizing what shipped.
pack-submit
Package one of this agent's own skills as a standalone community pack and submit it to the aeon registry as a PR.
updater_guide
Guidance for checking for and installing Row-Bot updates.