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 skills add IchenDEV/prompt-optimizer-plugins --skill select-claude-modelsgit clone --depth 1 https://github.com/IchenDEV/prompt-optimizer-pluginsWrote 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/skills/ichendev/prompt-optimizer-plugins/select-claude-models)<a href="https://agentmods.dev/skills/ichendev/prompt-optimizer-plugins/select-claude-models"><img src="https://agentmods.dev/badge/skills/ichendev/prompt-optimizer-plugins/select-claude-models/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/skills/ichendev/prompt-optimizer-plugins/select-claude-models"><img src="https://agentmods.dev/badge/skills/ichendev/prompt-optimizer-plugins/select-claude-models.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 | $0.00080 | $0.01559 |
| Opus 5.5 | $0.00032 | $0.00624 |
| Sonnet 5.5 | $0.00016 | $0.00312 |
| Haiku 4.5 | $0.00008 | $0.00156 |
Grade A, and why
select-claude-models 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 19d 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Select Claude Models
Help the user pick Claude models for a workload, especially multi-agent setups where a stronger model plans or advises and a faster model executes. Preserve their cost, latency, and quality constraints. Prefer Anthropic's measured patterns over inventing a custom hierarchy.
Default multi-agent recommendation for most coding and agent products in the Claude 5.5 family: Opus 想，Sonnet 干活 — Claude Opus 5.5 for judgment, planning, and hard decisions; Claude Sonnet 5.5 for execution, iteration, and bulk tool work.
Use the official basis
Read references/official-guidance.md for the selection matrix, advisor versus orchestrator patterns, effort defaults, pairing rules, and copy-ready executor prompt blocks. Treat its URLs as canonical and its prose as a dated working summary. Fetch the live Anthropic pages before making claims about current model IDs, prices, compatibility, or beta headers.
When the user then needs prompt text for a chosen model, hand off to the matching optimizer skill (optimize-claude-sonnet-5-5-prompts, optimize-claude-opus-5-prompts, or optimize-claude-fable-5-prompts).
Follow the workflow
1. Capture the workload contract
Identify:
- the task shape: everyday coding, polished docs, repeated agent jobs, long-horizon autonomy, deep research, high-volume low-latency, or mixed;
- whether success is checkable (tests, build, rubric) or judgment-heavy;
- latency sensitivity and budget sensitivity;
- whether the harness already supports tools, subagents, Managed Agents, or the advisor tool;
- whether the user wants a single model, an advisor-executor pair, or an orchestrator-worker team;
- any hard constraints on which models they can call.
Treat these as diagnostic dimensions. Do not invent product architecture the user did not ask for.
2. Apply the clarity gate
Ask only when the answer would change the model map. Blocking examples:
- "帮我选模型" with no task description and no cost/latency preference;
- a request that could mean either advisor-executor or orchestrator-worker, and the harness supports only one;
- a requirement to stay under a budget that cannot be inferred from "cheaper" alone.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 19d ago First seen · 112 lines · 80 tokens per session scan A 80145c8ef17e
select-claude-models is a skill published in the GitHub repository IchenDEV/prompt-optimizer-plugins (6 stars, last pushed 20d ago), licensed MIT. It adds 80 tokens to every session and 1,559 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-30.
Other skills, from other repositories
sol-terra-luna-auto-router
Optimize quality per token across GPT Codex, GPT Work, and Codex CLI by automatically selecting the best-fit GPT-5.6 Luna, Terra, or Sol profile and reasoning effort for the main task and every delegated subtask. Prioritize direct visible Codex execution and importance-aware subagent routing. Use automatically for…
watch
File sentinel that monitors the working directory for changes and marker comments, then auto-triggers appropriate skills. Poll-based via git diff against the last scan commit. Writes intake items for batch processing and routes marker actions through /do. Use for automatic reactions to file changes; do NOT use for…
pr-watch
Local PR watcher. Monitors CI status, automatically fixes failing checks by reading failure logs and applying targeted fixes, then optionally merges when all checks pass. Local CLI analog to Claude Code's cloud auto-fix feature.
live-preview
Mid-build visual verification loop. Takes screenshots of components during construction, not just after. Catches visual regressions and invisible features before they compound. Requires Playwright or similar screenshot tool.
review
5-pass structured code review — correctness, security, performance, readability, consistency.
marshal
Meta-orchestrator that takes any direction — broad, specific, or vague — and autonomously chains skills and context into actionable work. Gathers context from codebase, docs, and memory. Only asks the user when it genuinely cannot proceed. Single-session orchestrator.