run-burn

run-burn is a skill for Claude Code from Jerry0022/dotclaude. It costs 132 tokens per session (3,626 once invoked), scanned A, original, MIT.

An explicitly activated workflow for running many autonomous tasks in parallel. It gathers tasks from available sources and starts an independent task runner.

In plain words
What is it for?
Use it only by explicitly invoking /run-burn when you want to collect and run multiple tasks with aggressive parallelization.
Why use it?
It is intended for high-throughput work when you deliberately want to use substantial remaining usage and accept less supervision.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: reads .claude/ paths; names the AskUserQuestion tool.

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 devops plugin — 25 skills, 12 agents, 5 hooks, 3 MCP servers shipped together

Good fit Use it only by explicitly invoking /run-burn when you want to collect…

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 Jerry0022/dotclaude
Claude Code
/plugin install devops

Made for: Claude Code.

Or install devops, the plugin that ships this one along with the rest of its 25 skills, 12 agents, 5 hooks, 3 MCP servers.

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 run-burn

README.md
[![agentmods](https://agentmods.dev/badge/skills/jerry0022/dotclaude/run-burn.svg)](https://agentmods.dev/skills/jerry0022/dotclaude/run-burn)
Your own site
<a href="https://agentmods.dev/skills/jerry0022/dotclaude/run-burn"><img src="https://agentmods.dev/badge/skills/jerry0022/dotclaude/run-burn.svg" alt="Measured on agentmods" height="20"></a>
Per session 132 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,626 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00132 $0.03626
Opus 5 $0.00066 $0.01813
Sonnet 5 $0.00026 $0.00725
Haiku 4.5 $0.00013 $0.00363

Measured today against content hash 4f1313960e7e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

run-burn 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 today.

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.

plugins/devops/skills/run-burn/SKILL.md · 342 lines

How it starts

The opening of the file, as written. The whole thing — 342 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Run Burn

Consume the remaining weekly token budget before it resets, and turn it into landed, verified work rather than spend rate. Collect tasks from all sources, derive a plan that is measurably more than a normal /run-agents run, then launch autonomous mode with that plan.

Burn spends along two dimensions — depth (better models, higher effort, extra review passes per task) and breadth (more parallel lanes). Depth is preferred because it raises the cost of a task without raising how many tasks can be lost when the limit hits; breadth fills the remaining throughput gap. The derivation, the uplift floor that makes a burn worth running, the reserve, and the durability protocol all live in deep-knowledge/burn-scheduler.mdread that file at the start of Step 2.

This skill MUST only run when explicitly invoked via /run-burn. Never trigger from hooks, prompt phrasing, or heuristic matching.

Step 0 — Load Extensions

Silently check (do not surface "not found"):

  1. ~/.claude/skills/run-burn/SKILL.md + reference.md
  2. {project}/.claude/skills/run-burn/SKILL.md + reference.md
  3. Merge: project > global > plugin defaults

Extension-overridable knobs: RESERVE, LANE_CAP, BASE_PCT_PER_LANE_HOUR (see deep-knowledge/burn-scheduler.md).

Step 0.5 — Resume Detection

Before the confirmation gate, check for BURN-STATE.json in the project root.

If it exists with a non-empty queue or inFlight, a previous burn was cut off — by the weekly limit, a crash, or a session end. Ask via AskUserQuestion:

Ein abgebrochener Burn liegt vor: {done} Tasks gelandet, {queue} offen, {inFlight} unklar. Fortsetzen oder neu starten?

Options: ["Fortsetzen", "Neu starten"]

  • Fortsetzen → adopt integrationBranch, skip everything in done, verify each inFlight branch for commits and either merge it or requeue the task. Then run Step 2 (the stored plan is from the previous window and is stale — profile and lanes must be re-derived from current usage), skip Steps 3–5 (the queue already exists), and continue at Step 6.
  • Neu starten → archive to BURN-STATE.prev.json (never delete — it is the only record of the previous run's branches) and continue to Step 1.

Read the full file on GitHub · 342 lines

Files

What ships with it

2 files 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.

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. today Changed · +104 lines · +44 tokens per session 4f1313960e7e
  2. 6d ago First seen · 238 lines · 88 tokens per session scan A bf58e5b7722e

Subscribe to this mod's changes

run-burn is a skill published in the GitHub repository Jerry0022/dotclaude (5 stars, last pushed today), licensed MIT. It adds 132 tokens to every session and 3,626 once invoked, about $0.0007 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 2026-08-31.