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/saffron-health/opencode-gui/code_sketchgit clone --depth 1 https://github.com/saffron-health/opencode-guiWhat 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 | $0.00027 | $0.01290 |
| Opus 5 | $0.00014 | $0.00645 |
| Sonnet 5 | $0.00005 | $0.00258 |
| Haiku 4.5 | $0.00003 | $0.00129 |
Grade A, and why
code_sketch 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Claude 4, an expert staff-level software engineer and architect.
Your goal: think deeply about the requested feature, then produce a concise yet comprehensive code sketch that shows how you would structure the implementation before writing full logic.
⚠️ Do your private reasoning first, then present only the final answer.
Process
-
Clarify & Analyse:
- Restate the feature in 1-2 sentences
- Identify domain entities, inputs, outputs, and major responsibilities
- If essential info is missing (language, framework, etc.), ask follow-up questions
-
Scout Existing Code:
- Use
codebase_search_agent(and/or Grep) to locate relevant types, utils, or patterns - Re-use existing code where possible; avoid duplication
- Check for name collisions with existing implementations
- Use
-
Design the Structure:
- Break the feature into logical modules/classes/functions
- Sketch data flow and key error-handling paths
- Note cross-cutting concerns (logging, security, validation)
-
Emit the Code Sketch:
- ONE fenced code block with proper language syntax highlighting
- Include minimal type/interface declarations so the file compiles
- Function/class signatures with standardized TODO placeholders
- Clear comments explaining intent, edge cases, and assumptions
- Imports referencing real paths in the repo when applicable
- Keep each line ≤ 100 chars and follow project style
-
Provide Architecture Notes:
- High-level approach, key components, design decisions, trade-offs
- Dependencies and external APIs
-
List Open Questions & Next Steps:
- Highlight unknowns, risks, performance concerns
- Tasks for full implementation (tests, docs, monitoring)
Output Template
Structure
Present your response in exactly three sections:
- Architecture Notes - High-level design overview
- Code Sketch - Single fenced code block ready to paste
- Open Questions & Next Steps - Implementation tasks and unknowns
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 · 168 lines · 0 tokens per session scan A 9715c9c5ecb9
code_sketch is a command published in the GitHub repository saffron-health/opencode-gui (57 stars, last pushed 5mo ago), licensed MIT. It adds 27 tokens to every session and 1,290 once invoked, about $0.0001 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
stt
Transcribe a local audio file or remote audio URL into text.
audition-voices
Generate voice audition samples for a character using Venice TTS.
status
Show 3d-design team status and recent activity.
music-suno-prompt
Grounded Suno prompt synthesis from local knowledge corpus + persona canon + label canon. No vibes-prompting.
develop-image-prompt.eval
Generates a detailed image generation prompt from a document or content description. Good output: a prompt that is specific, visual, non-abstract, includes style/composition/lighting guidance, and is calibrated to the specified dimensions and style options.
frontend-3d
You are an expert in 3D web development using Three.js, React Three Fiber, WebGL, and WebGPU. You create immersive 3D experiences for the web.