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
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/nyldn/claude-octopus/model-configgit clone --depth 1 https://github.com/nyldn/claude-octopusWrote 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/model-config)<a href="https://agentmods.dev/commands/nyldn/claude-octopus/model-config"><img src="https://agentmods.dev/badge/commands/nyldn/claude-octopus/model-config.svg" alt="Measured on agentmods" 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 | $0.00009 | $0.05724 |
| Opus 5 | $0.00005 | $0.02862 |
| Sonnet 5 | $0.00002 | $0.01145 |
| Haiku 4.5 | $0.00001 | $0.00572 |
Grade B, and why
model-config scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
helper="$(find "${HOME}/.claude/plugins/cache" "${HOME}/Library/Application Support/Claude" "${LOCALAPPDATA:-/dev/null}/Claude" "${XDG_DATA_HOME:-${HOME}/.local/share}/Claude" -maxdepth 8 -path "*/nyldn-plugins/octo/*/sc Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
printf "ollama:%s\n" "$(command -v ollama >/dev/null 2>&1 && curl -sf --connect-timeout 1 --max-time 3 http://localhost:11434/api/tags >/dev/null 2>&1 && echo running || command -v ollama >/dev/null 2>&1 && echo installe How it starts
The opening of the file, as written. The whole thing — 538 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model Configuration
Your first output line MUST be: 🐙 Octopus Model Config
STEP 0: Emit Banner (MANDATORY — run before AskUserQuestion or any other step)
echo "🐙 Octopus Model Config"
echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
echo "provider | model | config | routing | cost"
Preflight — Ensure plugin root is resolvable (run via Bash tool):
OCTO_ROOT="${HOME}/.claude-octopus/plugin"
if [[ ! -x "$OCTO_ROOT/scripts/orchestrate.sh" ]]; then
helper="$OCTO_ROOT/scripts/helpers/ensure-plugin-root.sh"
if [[ ! -x "$helper" ]]; then
helper="$(find "${HOME}/.claude/plugins/cache" "${HOME}/Library/Application Support/Claude" "${LOCALAPPDATA:-/dev/null}/Claude" "${XDG_DATA_HOME:-${HOME}/.local/share}/Claude" -maxdepth 8 -path "*/nyldn-plugins/octo/*/scripts/helpers/ensure-plugin-root.sh" -print -quit 2>/dev/null)"
fi
[[ -x "$helper" ]] && bash "$helper" >/dev/null 2>&1 || true
fi
test -x "$OCTO_ROOT/scripts/orchestrate.sh" && echo "plugin-root:ok" || echo "plugin-root:missing"
If the output is plugin-root:missing, stop and ask the user to run /octo:setup.
Run this unconditionally — even when arguments are provided or when going to interactive wizard. The explicit bash block ensures the banner emits even when the command routes straight to AskUserQuestion (which historically skipped the inline-prose instruction and broke E2E pattern matching — see #301).
Interactive model configuration wizard. Detects installed providers, shows current settings, and guides users through configuration with AskUserQuestion.
STEP 1: Detect & Display
Run a SINGLE comprehensive detection command:
echo "=== Provider Detection ==="
printf "codex:%s\n" "$(command -v codex >/dev/null 2>&1 && echo installed || echo missing)"
printf "perplexity:%s\n" "$([ -n "${PERPLEXITY_API_KEY:-}" ] && echo configured || echo missing)"
printf "openrouter:%s\n" "$([ -n "${OPENROUTER_API_KEY:-}" ] && echo configured || echo missing)"
printf "copilot:%s\n" "$(command -v copilot >/dev/null 2>&1 && echo installed || echo missing)"
printf "qwen:%s\n" "$(command -v qwen >/dev/null 2>&1 && echo installed || echo missing)"
printf "kimi:%s\n" "$(command -v kimi >/dev/null 2>&1 && echo installed || echo missing)"
printf "ollama:%s\n" "$(command -v ollama >/dev/null 2>&1 && curl -sf --connect-timeout 1 --max-time 3 http://localhost:11434/api/tags >/dev/null 2>&1 && echo running || command -v ollama >/dev/null 2>&1 && echo installed || echo missing)"
printf "opencode:%s\n" "$(command -v opencode >/dev/null 2>&1 && echo installed || echo missing)"
echo "=== Config ==="
if [[ -f ~/.claude-octopus/config/providers.json ]]; then
cat ~/.claude-octopus/config/providers.json
else
echo "NO_CONFIG"
fi
echo "=== Env ==="
env | grep -E '^OCTOPUS_|^CLAUDE_MODEL=' 2>/dev/null || echo "none"
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 First seen · 538 lines · 9 tokens per session scan B fed487ec8237
model-config is a command published in the GitHub repository nyldn/claude-octopus (4,045 stars, last pushed today), licensed MIT. It adds 9 tokens to every session and 5,724 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other commands, from other repositories
papi-audit
Audit the latest changes in this branch for bugs and compliance with the project's conventions defined in CLAUDE.md.
test
You are filling in a test finding for this session. Read the docs/ directory for any test schema or guidelines first.
publish
Release a new package version — semver bump from commits, tag, registry publish, GitHub Release.
bump
Full dependency and toolchain bump — deps to latest major, Node pins, GitHub Actions, audit, validate, commit.
rebase
Rebase the current branch on main, a local branch, or a remote branch (with fetch).
toh-help
Display all Toh Framework commands and quick usage guide.