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.
git 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/prd-score)<a href="https://agentmods.dev/commands/nyldn/claude-octopus/prd-score"><img src="https://agentmods.dev/badge/commands/nyldn/claude-octopus/prd-score.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 5.1 | $0.00014 | $0.01081 |
| Opus 5 | $0.00007 | $0.00541 |
| Sonnet 5 | $0.00003 | $0.00216 |
| Haiku 4.5 | $0.00001 | $0.00108 |
Grade A, and why
prd-score 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 yesterday.
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.
This is a copy
94% identical to octo-prd-score — 7 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
STOP - DO NOT INVOKE /skill OR Skill() AGAIN
This command is already executing. The PRD file to score is: $ARGUMENTS.file
Instructions
Score the PRD against the 100-point AI-optimization framework.
Step 1: Load the PRD
Read the file at $ARGUMENTS.file using the Read tool.
Step 2: Evaluate Against Framework
Score each category:
Category A: AI-Specific Optimization (25 points)
- Sequential Phases: 0-10 pts (phases ordered by dependencies, each 5-15 min work)
- Explicit Non-Goals: 0-8 pts (dedicated Non-Goals section with explicit boundaries)
- Structured Format: 0-7 pts (FR codes, consistent headings, Given-When-Then criteria)
Category B: Traditional PRD Core (25 points)
- Problem Statement: 0-7 pts (quantified pain points, metrics)
- Goals & Metrics: 0-8 pts (SMART goals, P0/P1 priorities)
- User Personas: 0-5 pts (named personas with scenarios)
- Technical Specs: 0-5 pts (architecture, integrations, data models)
Category C: Implementation Clarity (30 points)
- Functional Requirements: 0-10 pts (FR codes, P0/P1/P2, acceptance criteria)
- Non-Functional Requirements: 0-5 pts (security, performance, reliability)
- Architecture: 0-10 pts (diagrams, data flow, API contracts)
- Phased Implementation: 0-5 pts (clear phases, time estimates, deliverables)
Category D: Completeness (20 points)
- Risk Assessment: 0-5 pts (3-5 risks with mitigations)
- Dependencies: 0-3 pts (external and internal)
- Examples: 0-7 pts (code snippets, API examples)
- Documentation Quality: 0-5 pts (formatting, ToC, glossary)
Step 3: Generate Score Report
Output:
## PRD Score Report: [PRD Title]
### Overall Score: XX/100 ([Grade])
Grade Scale: A+ (90-100), A (80-89), B (70-79), C (60-69), D (<60)
| Category | Score | Max |
|----------|-------|-----|
| A. AI-Specific Optimization | XX | 25 |
| B. Traditional PRD Core | XX | 25 |
| C. Implementation Clarity | XX | 30 |
| D. Completeness | XX | 20 |
### Top 3 Improvement Recommendations
1. [Highest impact fix] - +X points
2. [Second priority] - +X points
3. [Third priority] - +X points
### Verdict
[1-2 sentence summary of PRD quality and AI-readiness]
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.
- yesterday First seen · 124 lines · 14 tokens per session scan A f22da5be35ae
prd-score is a command published in the GitHub repository nyldn/claude-octopus (4,050 stars, last pushed today), licensed MIT. It adds 14 tokens to every session and 1,081 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 94% identical to octo-prd-score, differing in 7 lines, and is treated as a copy.
Other commands, from other repositories
composite-actions
Generate, review, secure, and test composite GitHub Actions following best practices — full repo scaffold, interview-driven generation, PR creation on existing repos, SHA pinning, secrets-as-inputs, job summaries, and actionlint validation.
linkerd
Linkerd-specific diagnostics — mTLS verification, proxy injection issues, authorization policy debugging, traffic management, and multi-cluster connectivity problems.
standup
Show a daily standup summary with completed, in-progress, and blocked tasks across all active epics.
feat
Kick off tasks creation for a new feature (4-item interview before invoking pm-agent; an epic-sized requirement routes to phased decomposition, a feature already built on one platform routes to cross-platform parity, a change that touches no endpoint routes to client-only).
close-sprint
Close the sprint — record Increment + velocity, return unfinished stories/bugs to the backlog.
workspace-status
(Workspace) Dashboard across all managed repos — sprint, unreleased, blocked, last sync.