start-issue

start-issue is a skill for Claude Code from cboone/agent-harness-plugins. It costs 51 tokens per session (7,803 once invoked), scanned C, original, MIT.

A workflow skill that creates a Git worktree, branch, and tmux window for a GitHub issue, then starts a separate session to plan the work. A GitHub issue is a tracked task or problem in a repository.

In plain words
What is it for?
Use it to start work on one or more GitHub issues in separate worktrees, with optional approval before execution.
Why use it?
It separates issue work from the current directory and branch, reducing setup work and keeping the planning session isolated.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions CLAUDE.md; mentions Claude Code; mentions AGENTS.md.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the workmux plugin — 2 skills shipped together

Good fit Use it to start work on one or more GitHub issues in separate worktrees, with optional approval before execution.

Compare 6 skills from other repositories ↓
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 cboone/agent-harness-plugins
Claude Code
/plugin install workmux

Made for: Claude Code.

Or install workmux, the plugin that ships this one along with the rest of its 2 skills.

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 start-issue

README.md
[![agentmods](https://agentmods.dev/badge/skills/cboone/agent-harness-plugins/start-issue/github.svg)](https://agentmods.dev/skills/cboone/agent-harness-plugins/start-issue)
Your own site
<a href="https://agentmods.dev/skills/cboone/agent-harness-plugins/start-issue"><img src="https://agentmods.dev/badge/skills/cboone/agent-harness-plugins/start-issue/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 start-issue

Your own site · 80×15
<a href="https://agentmods.dev/skills/cboone/agent-harness-plugins/start-issue"><img src="https://agentmods.dev/badge/skills/cboone/agent-harness-plugins/start-issue.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 51 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,803 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00051 $0.07803
Opus 5.5 $0.00020 $0.03121
Sonnet 5.5 $0.00010 $0.01561
Haiku 4.5 $0.00005 $0.00780

Measured 4d ago against content hash 0016481f9f41, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade C, and why

start-issue scanned grade C with 1 finding 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 4d 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.

Tells the agent never to refusehighAnti-refusal

Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.

**Never refuse.** If the user says to proceed, proceed: they always have a reason, and taking a claim over is recorded in step 6.
plugins/workmux/skills/start-issue/SKILL.md · 410 lines

How it starts

The opening of the file, as written. The whole thing — 410 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Address Issue in Worktree

Create a worktree, branch, and tmux window for a GitHub issue, then have the new session run address-issue to plan the work and stop for approval.

The approval gate lives in address-issue, not here. This skill creates the worktree and injects a prompt telling the new session which command to run; the plan is produced and approved in that session, not in this one.

To create the worktree without chaining into address-issue, use workmux:create-worktree instead.

Options

The user may provide these options inline:

  • --no-approval: Pass --no-approval through to the chained address-issue command, so the new session plans and executes without stopping
  • --resource <name>: Claim a named exclusive resource for the new worktree, and report the holder first if one holds it already
  • --fan-out: With several issue numbers, open one worktree, branch, tmux window and session per issue instead of one for all of them. --each and --separate are aliases with identical behavior; treat all three as the same fan-out mode before checking how it combines with other options

To list or release claims without creating a worktree, use workmux:create-worktree, which carries --list-resources and --release-resource.

Several Issues

Several issue numbers, as in /workmux:start-issue 42 57, have two readings, and each is a defined mode rather than a judgment call:

  • Combined (default). One worktree, branch, tmux window and session for all the issues together, as one piece of work. Run the workflow below once, with the changes listed under "Combined mode".
  • Fan-out (--fan-out, --each, --separate). One worktree per issue, each running /address-issue N for its own issue. Follow "Fan-Out Workflow" instead of the single-issue workflow.

The user knows which reading they mean when they type the command, so the flag decides; do not ask which mode to use. Count distinct issues before checking option interactions. If only one remains, any of the three flags is a no-op: follow the single-issue workflow, including --resource.

Read the full file on GitHub · 410 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. 4d ago First seen · 410 lines · 51 tokens per session scan C 0016481f9f41

Subscribe to this mod's changes

start-issue is a skill published in the GitHub repository cboone/agent-harness-plugins (2 stars, last pushed today), licensed MIT. It adds 51 tokens to every session and 7,803 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-13.

Related

Other skills, from other repositories