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 agentmods add skills/modem-dev/hunk/hunk-releasenpx skills add modem-dev/hunk --skill hunk-releasegit clone --depth 1 https://github.com/modem-dev/hunkWhat 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 | $0.00038 | $0.02360 |
| Opus 5 | $0.00019 | $0.01180 |
| Sonnet 5 | $0.00008 | $0.00472 |
| Haiku 4.5 | $0.00004 | $0.00236 |
Grade A, and why
hunk-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 — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunk release workflow
Maintainer-focused and source-checkout only. The tag workflow publishes hunkdiff plus five platform packages, attests the binary archives, and creates the GitHub release.
Safety
- Ask when the version, release branch, previous tag, or channel is ambiguous.
- Get explicit confirmation before pushing a tag, triggering publication, or editing a public release.
- Never reuse an npm version or move a tag after publication.
- Never bypass a benchmark regression without an approved, recorded reason.
- Never retry a partial publish before inventorying every package and artifact.
1. Confirm the release
Record the version, tag, branch, previous tag, and expected channel:
- newer stable: npm
latestand GitHub Latest; - prerelease: npm
beta, not GitHub Latest; - older-series backport: npm
backport-X.Y, leaving both latest pointers unchanged.
For backports, include only commits present between the previous tag and the release tip on that maintenance branch. Inspect .changeset/pre.json before changing prerelease state.
2. Prepare
Start from a clean, current release branch:
git status --short --branch
git fetch origin --tags --prune
git tag --sort=-version:refname | head -10
bun install --frozen-lockfile
bun run changeset:status
Read the pending Changesets, then generate metadata and the published release notes:
bun run release:version
bun run generate:changelog
git diff -- package.json packages CHANGELOG.md .changeset website
Verify the intended versions, consumed Changesets, and new changelog section. Do not force a bump by hand-editing generated versions.
generate:changelog projects CHANGELOG.md into hunk.dev/changelog. Never hand-edit its output under website/src/content/docs/changelog/, website/public/changelog/, website/releases/dates.json, or website/releases/latest.json. For a minor or major release, add the hand-authored parts to website/releases/notes.json under the major.minor key and regenerate:
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 · 230 lines · 38 tokens per session scan A 98045bbfff5c
hunk-release is a skill published in the GitHub repository modem-dev/hunk (8,933 stars, last pushed yesterday), licensed MIT. It adds 38 tokens to every session and 2,360 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
revdiff
Review diffs, files, and documents with inline annotations in a TUI overlay, or answer questions about revdiff usage, configuration, themes, and keybindings. Opens revdiff in tmux/zellij/herdr/kitty/wezterm/cmux/ghostty/iterm2/emacs-vterm, captures annotations, and addresses them. Works in git, hg, and jj repos…
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
critique
Git diff viewer. Renders diffs as web pages, images, and PDFs with syntax highlighting. Use this skill when working with critique for showing diffs, generating diff URLs, or selective hunk staging.
agent-code-analyzer
Agent skill for code-analyzer - invoke with $agent-code-analyzer.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.