Claude Octopus is an orchestration project that sends research, design, and coding tasks to Claude Code and other AI model providers so their results can be compared. Developers use it for multi-model work, disagreement detection, reviews, persistent context, and an optional workflow that moves from discovery through delivery. The catalogue entries are its commands, skills, agents, instructions, hooks, plugins, and settings.
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.
/plugin marketplace add nyldn/claude-octopus/plugin install octoWrote 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.
[](https://agentmods.dev/commands/nyldn/claude-octopus/auto)<a href="https://agentmods.dev/commands/nyldn/claude-octopus/auto"><img src="https://agentmods.dev/badge/commands/nyldn/claude-octopus/auto/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.
<a href="https://agentmods.dev/commands/nyldn/claude-octopus/auto"><img src="https://agentmods.dev/badge/commands/nyldn/claude-octopus/auto.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00011 | $0.03831 |
| Opus 5 | $0.00005 | $0.01916 |
| Sonnet 5 | $0.00002 | $0.00766 |
| Haiku 4.5 | $0.00001 | $0.00383 |
Grade A, and why
auto scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
printf "ollama:%s\n" "$(command -v ollama >/dev/null 2>&1 && curl -sf http://localhost:11434/api/tags >/dev/null 2>&1 && echo available || echo missing)" This is a copy
97% identical to octo-auto — 8 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 345 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Smart Router (/octo:auto)
Single entry point for all Claude Octopus workflows. Analyzes your natural language request and routes to the optimal workflow automatically.
/octo:auto research OAuth authentication patterns
/octo:auto debug the failing test in auth.ts
/octo:auto should we use Redis or Memcached?
/octo:auto write tests for the payment module
/octo:auto create a complete e-commerce platform
All /octo:* commands also work directly, bypassing the router.
Premium automatic peer check
When Premium mode is active, eligible single-owner implementation, review, design, copywriting, and general work receives one automatic cross-provider peer check after the owner returns a result. Users do not need a flag or a second command. Budget and Standard modes do not add this call.
Octopus skips the extra check for quick or direct requests, setup and status
work, image and research routes, and workflows that already run councils,
debates, crossfire, parallel work, or a full review. Explicit provider/model
pins and provider restrictions still win. Set OCTOPUS_PREMIUM_PEER_CHECK=off
to disable only this automatic addition.
The peer receipt records whether the bounded result review completed. A peer response is advisory and does not prove correctness.
EXECUTION CONTRACT (Mandatory)
When the user invokes /octo:auto <query> or says octo <query>, you MUST use the native path below. The detailed routing table later in this file is reference material for the shared runtime, not a second classifier.
NATIVE PATH: SINGLE AUTHORITATIVE ROUTER
For every non-meta request, pass the complete original query to the shared automatic runtime before doing any local intent matching:
OCTO_ROOT="${CLAUDE_PLUGIN_ROOT:-${HOME}/.claude-octopus/plugin}"
bash "${OCTO_ROOT}/scripts/orchestrate.sh" auto "<full original query>"
Treat the runtime's task classification and selected workflow as authoritative. If it reports Direct mode, continue handling the original request natively in the current Claude Code conversation. If it dispatches an external or parallel workflow, let that workflow complete and do not start a second route. If it reports a Native workflow request, load the named installed command and follow that command's contract in the current conversation. Do not continue to STEP 1 or choose a different route. This avoids two classifiers choosing different workflows and keeps Premium peer policy at the workflow root.
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.
- today Changed · +60 lines 9f85e786ac22
- 3d ago Changed · +7 lines 3d4c67fcbb2b
- 7d ago First seen · 278 lines · 11 tokens per session scan A d42c47a44b9c
auto is a command published in the GitHub repository nyldn/claude-octopus (4,061 stars, last pushed today), licensed MIT. It adds 11 tokens to every session and 3,831 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). It is 97% identical to octo-auto, differing in 8 lines, and is treated as a copy.
Other commands, from other repositories
arrange-workspace-flow
Workflow for arranging a workspace: layout, reference source code, business/technical context, ecosystem setup.
api-aqa-flow
Workflow for backend API test automation: TMS / Issue Tracker test cases → automated API tests, HITL-gated.
research-flow
Workflow for deep project research with grounded references, parallel exploration, etc.
self-help-flow
Rosetta self-help: explains capabilities and usage. DEPRECATED.
plan
Draft and lead-review a plan for a goal and present it for approval, without starting any work.
OPSX: Archive
Archive a completed change in the experimental workflow.