create-release

A release-note preparation workflow for BC-Bench, a benchmark used to evaluate coding agents. It reviews changes since the previous version and drafts notes for maintainer review.

In plain words
What is it for?
Use it when preparing a BC-Bench release, reviewing merged pull requests, or recording the agent-harness and development-tool versions included in a release.
Why use it?
It reduces the manual work of deciding which merged changes could affect evaluation results and matching the project's existing release-note style.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/microsoft/bc-bench/create-release
Any agent
npx skills add microsoft/BC-Bench --skill create-release
Clone the repo
git clone --depth 1 https://github.com/microsoft/BC-Bench

Made for: Claude Code, Codex.

Per session 56 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 683 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00056 $0.00683
Opus 5 $0.00028 $0.00342
Sonnet 5 $0.00011 $0.00137
Haiku 4.5 $0.00006 $0.00068

Measured 2d ago against content hash dd83645dbb45, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

create-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 2d 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.

.github/skills/create-release/SKILL.md · 35 lines

How it starts

The opening of the file, as written. The whole thing — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Create Release

Prepare release notes for maintainer review. Do not change versions, commit, tag, push, or create a GitHub release unless explicitly asked.

Workflow

  1. Establish the range: Read the version from pyproject.toml and use v<version> as the target tag. If that tag exists, use it as the target ref and select the preceding semantic-version tag. Otherwise, use HEAD as the target ref and select the latest reachable semantic-version tag. Ignore non-version tags such as evaluation-run tags. Exclude uncommitted changes and stop if the range is empty, ambiguous, or not fully available locally.
  2. Study the format: Fetch and read the bodies of the three most recently published releases from microsoft/BC-Bench. Match their concise paragraph style, spacing, and # Versions section; do not use GitHub's automatically generated notes as the draft.
  3. Screen merged PRs: Treat each commit in the complete <previous-tag>..<target-ref> range as a merged PR. Review its title, changed files, and PR description, inspecting the diff only as needed to determine whether it may affect evaluation results. Classify every PR as included or omitted, but do not produce a PR-by-PR summary.
  4. Select release-note content: Include only changes that may affect evaluation results or their comparability, such as dataset and gold-answer changes, evaluation logic, prompts, models, agent harnesses, tools, result schemas, and execution behavior. Combine related PRs into outcome-focused paragraphs. Omit formatting, linting, tests, documentation, refactoring, dependency maintenance, and CI housekeeping unless they change benchmark behavior or reproducibility. Call out compatibility boundaries explicitly.
  5. Collect shipped versions: Read the target ref, not the uncommitted worktree, and locate the exact pins for @github/copilot, @anthropic-ai/claude-code, and Microsoft.Dynamics.BusinessCentral.Development.Tools. Confirm duplicate pins agree. Never copy versions from an older release or infer them from commit messages.
  6. Validate the draft: Ensure every claim is supported by an included PR, every PR in the range was classified, omitted PRs have no plausible evaluation impact, the target version is correct, and the three shipped versions are exact. Stop and report conflicting pins, missing history, or uncertain evaluation impact instead of guessing.

Read the full file on GitHub · 35 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. 2d ago First seen · 35 lines · 56 tokens per session scan A dd83645dbb45

Subscribe to this mod's changes

create-release is a skill published in the GitHub repository microsoft/BC-Bench (56 stars, last pushed 4d ago), licensed MIT. It adds 56 tokens to every session and 683 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens