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 skills add Netcracker/qubership-ai-packages --skill change-description-authoringgit clone --depth 1 https://github.com/Netcracker/qubership-ai-packagesWrote 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/netcracker/qubership-ai-packages/change-description-authoring)<a href="https://agentmods.dev/skills/netcracker/qubership-ai-packages/change-description-authoring"><img src="https://agentmods.dev/badge/skills/netcracker/qubership-ai-packages/change-description-authoring/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.
<a href="https://agentmods.dev/skills/netcracker/qubership-ai-packages/change-description-authoring"><img src="https://agentmods.dev/badge/skills/netcracker/qubership-ai-packages/change-description-authoring.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00174 | $0.15305 |
| Opus 5.5 | $0.00070 | $0.06122 |
| Sonnet 5.5 | $0.00035 | $0.03061 |
| Haiku 4.5 | $0.00017 | $0.01530 |
Grade A, and why
change-description-authoring 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 — 906 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Authoring a change description
This skill governs what a change description says, in what order, and for whom. Wording, tense,
mood, sentence length, punctuation, and dialect belong to english-developer-style; load it too,
and let it own the sentences. Subject length limits, prefix grammar, and trailer syntax belong to
commitlint or the platform, not to a prompt.
A report of a problem nobody has fixed yet is a different artifact with a different reader, and it is
problem-report-authoring's: an issue, a feature request, a question, a support ticket, or a message
to a colleague. Its reader holds no diff, so it establishes the problem before anything else, and what
it may then say about the cause and the fix is that skill's to decide. Do not carry the mechanism of
the Why slot or the change of the What slot into one.
A change description is three artifacts written at once: the commit message, the pull request description, and the changelog entry. They are read by five readers, at five different moments, holding five different things. The rules below name the reader every time, because the same sentence is essential to one reader and noise to another, and the only way to decide is to ask whose question it answers.
1. The correction that matters most
Write for the readers who are not in the room. The reviewer can ask; the other four cannot. The
archaeologist holds a line of code years later, the on-call engineer holds a stack trace at 3 a.m.,
the upgrading user holds release notes, and the backporter holds a list of subjects. Each of them
has
one artifact and one search string, and a description that answers the reviewer's questions in the
pull request while leaving git log with a title has served the one reader who needed it least.
| Reader | Holds | Opens | Asks |
|---|---|---|---|
| R1 Reviewing maintainer | The diff | The pull request description | Was something wrong? Is this behavior right? How was it checked? What is left out? |
| R2 Archaeologist | A line of code, via git blame |
The commit message | Why does this exist? Why this approach? Which constraint forced it? |
| R3 On-call engineer | A stack trace or a log line, and a version | The changelog, then the commit a bisect reaches | Is this my incident? What triggers it? Am I on an affected version? Is there a workaround? |
| R4 Upgrading user | The version they run and the one they want | The release notes | What changed that I can observe? Is it compatible? What must I do? |
| R5 Backporter | A maintenance branch | Labels, trailers, and the subject list | Is this a fix? How severe? Which versions? Does it need another change first? |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 Changed · +9 lines 0ee7e1a3b55c
- 9d ago Changed · +332 lines bf27035c37b7
- 1mo ago Changed · +6 lines 6685c0374b26
- 1mo ago First seen · 559 lines · 174 tokens per session scan A cd3ee460024d
change-description-authoring is a skill published in the GitHub repository Netcracker/qubership-ai-packages (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 174 tokens to every session and 15,305 once invoked, about $0.0007 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 2026-09-05.
Other skills, from other repositories
changelog
Auto-generate a changelog from git commits and sprint data. Internal and player-facing versions.
comet-safe-delivery
A Chinese-language procedure for safely delivering specified Comet changes through Git. It covers checking worktrees and unrelated edits, staging exact files, validating hooks, and authorized commits or pushes.
close-task-commit-push-pr
Close the active backlog task (detected from branch name), commit all changes, push to remote, and open a pull request. Use when the user says "close task and ship it", "close task commit push pr", or invokes /close-task-commit-push-pr.
git-workflow
Guided git workflows: prepare PRs, clean up branches, resolve merge conflicts, handle monorepo tags, squash-and-merge patterns. Use when asked to prepare a PR, clean branches, resolve conflicts, or tag a release.
git-push
A GitHub publishing workflow that handles first-time uploads, routine updates, and tagged releases. Git is a system for recording code changes, while GitHub hosts repositories and releases online.
development-workflow
Detailed development workflow with modular patterns for git, review, testing, and deployment.