bifrost: Skill for Claude Code

.claude/skills/split-commit-into-stack/SKILL.md

split-commit-into-stack is a skill for Claude Code from maximhq/bifrost. It costs 96 tokens per session (3,693 once invoked), scanned A, original, Apache-2.0.

A workflow for breaking one oversized Git branch or commit into several independently reviewable Graphite pull requests. Each part should tell a coherent story and work on its own.

In plain words
What is it for?
Use it to measure a branch, choose sensible split points, create a stack of pull requests, and verify that each lower part builds and passes independently.
Why use it?
Large pull requests are difficult to review when unrelated decisions are mixed together.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions CLAUDE.md; names the AskUserQuestion tool; positional $N argument.

This is maximhq/bifrost's own configuration. It tells Claude Code how to work on bifrost 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 bifrost configures →

About the project

Bifrost is an AI gateway that gives applications one OpenAI-compatible API for accessing models from more than 23 providers. It is used to route model requests, handle provider failover and load balancing, and apply features such as caching, guardrails, and MCP gateway support.

maximhq/bifrost · 8,898 stars · on GitHub · getmaxim.ai

Reuse

Borrowing it

Nothing to install: this file belongs to maximhq/bifrost. 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/maximhq/bifrost/dev/.claude/skills/split-commit-into-stack/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/maximhq/bifrost

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 split-commit-into-stack

README.md
[![agentmods](https://agentmods.dev/badge/skills/maximhq/bifrost/split-commit-into-stack/github.svg)](https://agentmods.dev/skills/maximhq/bifrost/split-commit-into-stack)
Your own site
<a href="https://agentmods.dev/skills/maximhq/bifrost/split-commit-into-stack"><img src="https://agentmods.dev/badge/skills/maximhq/bifrost/split-commit-into-stack/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 split-commit-into-stack

Your own site · 80×15
<a href="https://agentmods.dev/skills/maximhq/bifrost/split-commit-into-stack"><img src="https://agentmods.dev/badge/skills/maximhq/bifrost/split-commit-into-stack.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 96 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,693 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Tool Misuse · line 158
    Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
    Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
  • high Tool Misuse · line 207
    Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
    Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
How audits are shown
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 $0.00096 $0.03693
Opus 5.5 $0.00038 $0.01477
Sonnet 5.5 $0.00019 $0.00739
Haiku 4.5 $0.00010 $0.00369

Measured 1mo ago against content hash 81cdbe0a315e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

split-commit-into-stack 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 1mo 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.

.claude/skills/split-commit-into-stack/SKILL.md · 299 lines

How it starts

The opening of the file, as written. The whole thing — 299 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Split Commit Into Stack

Turn one large commit into a stack where each PR is a coherent story that compiles and passes on its own. The work is mostly analysis: the mechanics are five commands, but choosing the cut and proving it holds is where the value is.

Most of the cost of a big PR is that a reviewer cannot tell which changes belong to which decision. A split earns its keep when it separates decisions, not when it merely reduces line count.

Step 0: Measure before proposing anything

Measure the whole branch against its base, not the tip against its parent: a branch that already holds several commits is split as one diff, and HEAD~1 would silently measure, inspect and later reset only the last of them.

base="$(git merge-base "$(gt parent)" HEAD)"   # the branch's base, whatever the commit count
git --no-pager log --oneline "$base"..HEAD     # several commits that already match the concerns: see Graphite notes
git --no-pager diff --numstat "$base" HEAD | awk '{i+=$1;d+=$2; printf "+%-6d -%-6d %s\n",$1,$2,$3} END {printf "TOTAL +%d -%d\n",i,d}'
git --no-pager diff --name-status "$base" HEAD

Every later step reads "$base" from this shell, so run them in the same one or set it again.

Two numbers change the conversation before any splitting is discussed:

  • Production vs test insertions. A 3,500-line diff where 1,000 lines are one new test file is not a 3,500-line review. Report the production-only figure; sometimes it dissolves the problem and no split is warranted.
  • Per-concern totals. Group files by concern (below) and total each. If one group is 85% of the diff and is atomic, splitting the remaining 15% is the entire available win
    • say so plainly rather than proposing a heroic three-way split.
  • Whether one file serves both concerns. Two or three such files is normal and Step 4 usually dissolves them; a dozen means the concerns are not actually separable and the honest answer is to leave the commit whole.

Use --name-status and not just --numstat: added (A) and deleted (D) files change what the git add commands must look like, and a deletion left in the wrong PR breaks the build in a way that is annoying to diagnose later.

Read the full file on GitHub · 299 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. 1mo ago First seen · 299 lines · 96 tokens per session scan A 81cdbe0a315e

Subscribe to this mod's changes

split-commit-into-stack is a skill published in the GitHub repository maximhq/bifrost (8,898 stars, last pushed yesterday), licensed Apache-2.0. It adds 96 tokens to every session and 3,693 once invoked, about $0.0004 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 2026-09-02.

Related

Other skills, from other repositories

validate-release-green

Run or inspect the local Release-Green observer for an exact base or combined PR candidate, separating hard failures, advisory drift and incomplete evidence. Use for maintainer preflight; full CI acceptance remains a separate requirement.

diegosouzapw/OmniRoute · 47 tokens

maintainer-ops

Use for vLLM Semantic Router issue intake, PR decisions, releases, and reviewed GitHub mutations performed on behalf of maintainers.

vllm-project/semantic-router · 32 tokens

release-notes

Create release notes for a new version tag. Gathers all commits, PRs, issues fixed, and breaking changes since a previous release. Creates the release notes markdown file, tags the repo, and pushes. Asks the user to confirm the base version to diff against.

agentic-community/mcp-gateway-registry · 59 tokens

create-milestone

Create a GitHub milestone for an upcoming release. Suggests the next version based on the latest release, gathers all merged PRs and closed issues since that release, presents a draft with two tables (Issues and PRs) for user approval, then creates the milestone and assigns all approved items.

agentic-community/mcp-gateway-registry · 64 tokens

thumbgate-protect

Inspect this repo's branch and release governance (protected branches, release rules, protected-file globs) and, only when the user explicitly approves, grant a scoped, time-limited exception so a protected-file edit or publish can proceed under audit. Reads posture via the getbranchgovernance MCP tool and records a…

IgorGanapolsky/ThumbGate · 165 tokens

release-notes

Generate release notes from git history and CHANGELOG. Formats for GitHub Release, Slack, or email. Use before tagging a release.

srnichols/plan-forge · 31 tokens