An automated pass through a GitHub issue backlog that checks open issues, sends ready work to coding agents, monitors their pull requests, and returns a summary. GitHub labels record the work so repeating the pass does not redo it.

Use it to run a scheduled or on-demand backlog shift, limit how many issues are dispatched, preview actions with a dry run, and keep dispatched pull requests moving.

It removes repeated manual triage and dispatch work from an issue-to-pull-request process. It also separates unclear or blocked issues from work that is ready to start.

shift — Skill for Claude Code by bengous/claude-code-plugins

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code

/plugin marketplace add bengous/claude-code-plugins

Claude Code

/plugin install github-flow

Made for: Claude Code.

Or install github-flow, the plugin that ships this one along with the rest of its 8 skills, 1 agent.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/bengous/claude-code-plugins/shift/github.svg)](https://agentmods.dev/skills/bengous/claude-code-plugins/shift)

Your own site

<a href="https://agentmods.dev/skills/bengous/claude-code-plugins/shift"><img src="https://agentmods.dev/badge/skills/bengous/claude-code-plugins/shift/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/bengous/claude-code-plugins/shift"><img src="https://agentmods.dev/badge/skills/bengous/claude-code-plugins/shift.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 64 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 781 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.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00064	$0.00781
Opus 5	$0.00032	$0.00391
Sonnet 5	$0.00013	$0.00156
Haiku 4.5	$0.00006	$0.00078

Measured 2d ago against content hash b14a353e0091, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-16, from the pricing page.

Security

Grade A, and why

shift 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-flow/skills/shift/SKILL.md · 44 lines

How it starts

The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Shift

One pass over the tracker with nobody watching. The state lives on GitHub, in labels, never in a file: a shift that runs twice does its work once.

Input

$ARGUMENTS

--max <n>: issues dispatched this pass. Default 2, for cost.
--dry-run: print each action in place of taking it. No comment, no label, no worker.

Protocol

Labels. gh label list --limit 100. Create the ones it does not list: needs-info, dispatched, blocked for issues, shift, needs-human for PRs.
```
gh label create needs-info --description "Missing evidence or acceptance criteria"
```
Triage. gh issue list --state open --json number,title,labels,createdAt. Drop the ones already labelled dispatched, blocked, needs-info or epic. Oldest first, bug before the rest. Read ${CLAUDE_PLUGIN_ROOT}/skills/triage/SKILL.md and apply its protocol to each: it verifies the issue against the current code and closes or comments as its verdict fixes. Read it, do not invoke it: github-flow:triage is a manual skill. unclear → label needs-info. valid → candidate.
Dispatch. The first --max candidates, in one call to github-flow:dispatch with their numbers.
Maintain. gh pr list --state open --label shift --json number,headRefName,mergeable,statusCheckRollup,reviewDecision,comments. Skip the ones labelled needs-human. A PR with a red check, mergeable: CONFLICTING, or a review comment nobody answered goes to one github-flow:issue-worker in fix <pr> mode, model: opus and isolation: worktree on the call. Count the PR's comments whose first line is  before dispatching: at two, label needs-human and leave the PR alone.
Record the fix. One comment per fix, written to a file outside the repository and posted with gh pr comment <pr> --body-file <tmp>. First line , then one line naming what the worker changed, or why it reported blocked. These are the comments step 4 counts; without them the 2-attempt bound never fires.
Digest. The report is the deliverable.

Read the full file on GitHub · 44 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.

2d ago First seen · 44 lines · 64 tokens per session scan A b14a353e0091

Subscribe to this mod's changes

shift is a skill published in the GitHub repository bengous/claude-code-plugins (4 stars, last pushed today), licensed MIT. It adds 64 tokens to every session and 781 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 2026-09-14.

Claude Code Git and workflow Project management automation developer-tools workflow

work

Dispatch a tracker ticket into an isolated worktree and drive it autonomously to an open pull request, then stop for review. Pass interactive to brainstorm the ticket with a human before planning it, or a sentence of editorial direction as a rider. Use when asked to work a ticket end-to-end.

mikestankavich/claude-ship-workflow · 0 tokens

batch

Dispatch a night's batch of Todo tickets — one worktree and one pull request each — and leave a morning summary. Run explicitly; never inferred.

mikestankavich/claude-ship-workflow · 31 tokens

release

Verify, write the CHANGELOG, cut a version — then hand the release sequence back.

hamr0/agentic-toolkit · 19 tokens

implement-issue

Drive a GitHub issue from branch to pull request — read the issue, create a feature branch, implement the change, run the project's build and tests, then commit, push, and open a PR. Language and build-tool agnostic. Use when asked to implement, fix, or resolve a GitHub issue by number.

alexmond/alexmskills · 69 tokens

ag-team-terminais

Protocolo para 2+ sessoes Claude interativas em terminais separados trabalharem no MESMO repo com objetivos distintos. Coordenador + workers, manifesto duravel em docs/ai-state/team/, mensagens nativas via ListAgents/SendMessage, worktree por worker. Modos: init | join | status | fechar. Para teammates dentro de UMA…

andregusman-raiz/a-gusman-claude · 86 tokens

ag-registrar-issue

Registra GitHub Issues para problemas nao resolvidos imediatamente. Recebe contexto via SendMessage de outros agents. Dedup, labels, severidade. Use when an agent finds a problem it cannot fix now.

andregusman-raiz/a-gusman-claude · 49 tokens