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
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/nyldn/claude-octopusnpx agentmods add commands/nyldn/claude-octopus/octo-guardWrote 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/octo-guard)<a href="https://agentmods.dev/commands/nyldn/claude-octopus/octo-guard"><img src="https://agentmods.dev/badge/commands/nyldn/claude-octopus/octo-guard/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/octo-guard"><img src="https://agentmods.dev/badge/commands/nyldn/claude-octopus/octo-guard.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 | $0.00502 |
| Opus 5 | $0.00005 | $0.00251 |
| Sonnet 5 | $0.00002 | $0.00100 |
| Haiku 4.5 | $0.00001 | $0.00050 |
Grade C, and why
octo-guard scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **Careful mode**: Warns before destructive Bash commands (rm -rf, DROP TABLE, git push --force, etc.) Copies of this mod
2 near-identical copies found in the catalogue:
- guard — 92% identical, 3 lines differ
- octo-guard — 89% identical, 9 lines differ
What it actually says
Guard Mode - Full Safety Activation
Instructions
When the user invokes /octo:guard <directory>, activate both careful mode and freeze mode in a single command.
Activation
- Activate careful mode (destructive command warnings):
_OCTO_SESSION_ID="${CLAUDE_CODE_SESSION_ID:-${CLAUDE_SESSION_ID:-$$}}"
echo "active" > "/tmp/octopus-careful-${_OCTO_SESSION_ID}.txt"
- Activate freeze mode (edit boundary enforcement):
freeze_dir="$(cd "$1" 2>/dev/null && pwd)" || freeze_dir="$1"
_OCTO_SESSION_ID="${CLAUDE_CODE_SESSION_ID:-${CLAUDE_SESSION_ID:-$$}}"
echo "${freeze_dir}" > "/tmp/octopus-freeze-${_OCTO_SESSION_ID}.txt"
What It Does
Guard mode combines both safety primitives:
- Careful mode: Warns before destructive Bash commands (rm -rf, DROP TABLE, git push --force, etc.)
- Freeze mode: Blocks Edit/Write operations outside the specified directory
This is the recommended safety configuration for focused work in sensitive codebases.
Usage
/octo:guard src/auth
/octo:guard ./packages/core
/octo:guard /absolute/path/to/module
After activation, confirm to the user:
🛡️ Guard mode activated:
⚠️ Careful: Destructive commands require confirmation
🔒 Freeze: Edits restricted to <resolved-path>
Use /octo:unfreeze to remove edit restriction.
Argument Required
If invoked without a directory argument, activate only careful mode and ask about freeze:
⚠️ Careful mode activated (destructive command warnings).
Which directory should I restrict edits to? Example: /octo:guard src/auth
When to Use
- Starting work in a production-adjacent codebase
- Debugging with maximum safety
- When you want both protections without two separate commands
See Also
/octo:careful— Activate only destructive command warnings/octo:freeze— Activate only edit boundary enforcement/octo:unfreeze— Remove freeze restriction
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.
- 9d ago First seen · 75 lines · 11 tokens per session scan C 1b01a83d620c
octo-guard is a command published in the GitHub repository nyldn/claude-octopus (4,062 stars, last pushed today), licensed MIT. It adds 11 tokens to every session and 502 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
ai-context
Generate, update, or audit AI IDE context files with AGENTS.md as the canonical shared context and tool-specific bridge files. Signal Gate principle — only what agents cannot discover: $ARGUMENTS.
sync
Analyze codebase and populate knowledge-base with conventions, patterns, and technical debt.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.