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/nodnarbnitram/claude-code-extensions/agent-from-docsgit clone --depth 1 https://github.com/nodnarbnitram/claude-code-extensionsWhat 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 | $0.00008 | $0.00975 |
| Opus 5 | $0.00004 | $0.00487 |
| Sonnet 5 | $0.00002 | $0.00195 |
| Haiku 4.5 | $0.00001 | $0.00097 |
Grade A, and why
agent-from-docs 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 2d 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.
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.
How it starts
The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create Agent from Documentation
You will create a new specialized agent by analyzing documentation from the provided URLs.
Documentation URLs: $ARGUMENTS
Your Task
Follow these steps to create the agent:
Step 1: Analyze Documentation
If multiple URLs are provided, launch multiple technical-researcher agents in parallel (one per URL). If only one URL is provided, use a single researcher.
For parallel execution with multiple URLs:
- Launch all technical-researcher agents simultaneously in a single message with multiple Task tool calls
- Each agent should analyze its assigned URL independently
- Combine all research findings before proceeding to Step 2
Each researcher should investigate:
- What technology/framework/library is being documented
- Key concepts, APIs, methods, and patterns
- Best practices and conventions
- Common workflows and use cases
- Important implementation details that an expert should know
- How this technology differs from alternatives
Request a comprehensive summary suitable for creating a specialized agent.
Step 1.5: Determine Agent Creation Strategy
After gathering all research findings, analyze whether to create one or multiple agents:
Create MULTIPLE agents when URLs represent distinct verticals/technologies:
- Different core APIs or services (e.g., Cloudflare Workers vs D1 vs R2)
- Separate deployment models or runtime environments
- Distinct problem domains within a platform
- Independent tools/products that happen to share a brand
Create a SINGLE agent when URLs cover the same technology:
- Different aspects of one framework (e.g., Next.js routing + data fetching + deployment)
- Progressive documentation depth on one topic
- Multiple guides for the same library/tool
Examples:
- ✅ Multiple agents:
workers.cloudflare.com/docs+developers.cloudflare.com/d1+developers.cloudflare.com/r2→ 3 specialized agents - ❌ Single agent:
nextjs.org/docs/routing+nextjs.org/docs/api-routes+nextjs.org/docs/deployment→ 1 comprehensive agent
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.
- 2d ago First seen · 108 lines · 8 tokens per session scan A 79b728634a83
agent-from-docs is a command published in the GitHub repository nodnarbnitram/claude-code-extensions (16 stars, last pushed 4mo ago), licensed MIT. It adds 8 tokens to every session and 975 once invoked, about $0.0000 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 2026-08-30.
Other commands, from other repositories
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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
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.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.