jarvis: Skill for Claude Code

.claude/skills/jarvis-release/SKILL.md

jarvis-release is a skill for Claude Code from jarvis-intelligence/jarvis. It costs 144 tokens per session (2,316 once invoked), scanned C, original, MIT.

A release procedure for the Jarvis repository itself. It covers updating the version, recording the changes, merging a release pull request, creating a GitHub Release, and checking package publishing.

In plain words
What is it for?
Cutting a new Jarvis release after feature work is already merged, including version bumps, changelog updates, pull requests, tags, and publication checks.
Why use it?
It provides an ordered checklist for releasing the repository so version files, Git history, tags, and publishing steps stay aligned.

Skill for Claude Code

Written for Claude Code: Claude Code plugin machinery. Also seen: mentions Claude Code; mentions Codex.

This is jarvis-intelligence/jarvis's own configuration. It tells Claude Code how to work on jarvis 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 jarvis configures →

Reuse

Borrowing it

Nothing to install: this file belongs to jarvis-intelligence/jarvis. 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/jarvis-intelligence/jarvis/main/.claude/skills/jarvis-release/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/jarvis-intelligence/jarvis

Made for: Claude Code.

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 jarvis-release

README.md
[![agentmods](https://agentmods.dev/badge/skills/jarvis-intelligence/jarvis/jarvis-release/github.svg)](https://agentmods.dev/skills/jarvis-intelligence/jarvis/jarvis-release)
Your own site
<a href="https://agentmods.dev/skills/jarvis-intelligence/jarvis/jarvis-release"><img src="https://agentmods.dev/badge/skills/jarvis-intelligence/jarvis/jarvis-release/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 jarvis-release

Your own site · 80×15
<a href="https://agentmods.dev/skills/jarvis-intelligence/jarvis/jarvis-release"><img src="https://agentmods.dev/badge/skills/jarvis-intelligence/jarvis/jarvis-release.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 144 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,316 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 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 $0.00144 $0.02316
Opus 5 $0.00072 $0.01158
Sonnet 5 $0.00029 $0.00463
Haiku 4.5 $0.00014 $0.00232

Measured 10d ago against content hash 2ff6a2554022, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade C, and why

jarvis-release scanned grade C with 2 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 10d 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

curl -s https://pypi.org/pypi/<package-name>/json | python3 -c "import json,sys; print(sorted(json.load(sys.stdin)['releases'].keys()))"

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s https://pypi.org/pypi/<package-name>/json | python3 -c "import json,sys; print(sorted(json.load(sys.stdin)['releases'].keys()))"
.claude/skills/jarvis-release/SKILL.md · 130 lines

How it starts

The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Releasing jarvis

This captures jarvis's actual release process — every command here was run for real and verified working while cutting v0.3.1. Follow it in order; the steps are load-bearing on each other (CI gates the merge, the merge commit is what gets tagged, the tag is what the GitHub Release publishes, and publishing the release is what triggers PyPI/registry publishing).

Before starting: what's actually being released?

main must already contain everything you're about to release. If there's a pending feature/fix branch that hasn't been merged yet, merge that first (its own PR, its own CI, its own review) — the release step that follows is purely mechanical version-bumping on top of an already-correct main. Don't conflate "ship my feature" with "cut a release"; they're sequential, not the same PR.

1. Decide the version bump

Semver, and the CHANGELOG is explicit about the reasoning behind past bumps — read CHANGELOG.md's existing entries before picking, they set the calibration:

  • Patch (0.3.00.3.1): pure bug fixes, no new capability. v0.2.1 and v0.3.1 were both patches.
  • Minor (0.2.x0.3.0): new capability, even if delivered alongside fixes. v0.3.0's own CHANGELOG entry says "Minor rather than patch: Java/Kotlin repos are indexable for the first time, --search-only is a new mode..." — that annotation is the model to follow when it's a judgment call.
  • Major: hasn't happened yet in this repo's history; a breaking change to the MCP tool surface or CLI would warrant it.

If it's genuinely ambiguous (a fix that also quietly changes behavior), say what you're picking and why in one line before proceeding — this is the one place in the whole pipeline that isn't mechanical, and it's cheap to confirm before an irreversible PyPI upload locks it in.

2. Bump the version in all 3 files

These must all end up identical, or CI's tag/version guard (see step 6) fails the release:

File What to change
pyproject.toml version = "X.Y.Z"
server.json two fields: the top-level "version" and packages[0].version
uv.lock never hand-edit. Run uv lock — it updates this package's own self-referential version entry (Updated jarvis-mcp vX.Y.Z-1 -> vX.Y.Z in its output) and re-resolves nothing else changes if no deps moved

Read the full file on GitHub · 130 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. 10d ago First seen · 130 lines · 144 tokens per session scan C 2ff6a2554022

Subscribe to this mod's changes

jarvis-release is a skill published in the GitHub repository jarvis-intelligence/jarvis (0 stars, last pushed 2d ago), licensed MIT. It adds 144 tokens to every session and 2,316 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories