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 skills/tallesborges/zdx/model-dispatchnpx skills add tallesborges/zdx --skill model-dispatchgit clone --depth 1 https://github.com/tallesborges/zdxWhat 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 | $0.00071 | $0.01029 |
| Opus 5 | $0.00036 | $0.00515 |
| Sonnet 5 | $0.00014 | $0.00206 |
| Haiku 4.5 | $0.00007 | $0.00103 |
Grade A, and why
model-dispatch 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 3d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model Dispatch
Dispatch one prompt to one or more explicit models through zdx exec. This skill is loaded into the calling agent's context; it is not a subagent hop. Built-in subagents select predefined roles, while model dispatch selects named LLMs.
When to use
- Send work to one named model: "ask Opus to review this".
- Fan out one prompt to several models and show every answer.
- Compare the same model at multiple thinking levels.
- Compare exact model/thinking pairs, such as Opus at
lowversus GPT athigh. - Compare models inside one specialist role, such as three models answering as
oracle.
1. Define the runs
Each run is an explicit MODEL[@LEVEL][+SUBAGENT] spec passed with --run. The level is optional and defaults to medium; the subagent role is optional and defaults to none.
--run claude-cli:claude-opus-4-8
--run claude-cli:claude-opus-4-8@low
--run openai:gpt-5.5@high
--run openai:gpt-5.5@high+oracle
Supported levels are off, low, medium, high, xhigh, and max. Providers may map or clamp levels their models do not support exactly; mention this caveat when benchmarking reasoning levels.
Use the models the user named. Otherwise discover valid IDs with zdx models list or zdx models list --json and choose a small sensible set. Never invent model IDs.
2. Build the prompt
By default every run receives the full ZDX system prompt and project context. Still make the task self-contained: include the requested outcome, referenced source material, constraints, audience, and output shape.
Use --no-system-prompt only for an intentionally isolated comparison. Use --no-tools for clean one-shot answers. For prompts with quotes or newlines, write the prompt under $ZDX_ARTIFACT_DIR/tmp/ and pass it with --prompt-file.
To run the comparison in a specialist role instead of the default one, name a subagent (for example oracle or explorer). The role supplies its prompt, tools, and defaults, while the dispatched model and thinking level still win. Two ways to set it:
What ships with it
1 file 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.
- 3d ago First seen · 75 lines · 71 tokens per session scan A c80e8d4d2f87
model-dispatch is a skill published in the GitHub repository tallesborges/zdx (20 stars, last pushed 4d ago), licensed MIT. It adds 71 tokens to every session and 1,029 once invoked, about $0.0004 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 skills, from other repositories
context-fundamentals
This skill should be used to explain or reason about the foundational concepts of context engineering: what context is, the anatomy of a context window, how attention mechanics work, the U-shaped attention curve, why context quality matters more than quantity, and the mental models needed to interpret every other…
enhance-prompt
Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.
prompt-optimization
Improve a prompt on the evaluations workbench through a measured loop. Score the baseline first, then duplicate the target column, form a hypothesis from failing rows, edit the copy's prompt draft, run, compare pass rate and cost, and repeat until the numbers hold. Use when the user asks to optimize or improve a…
prompt-engineer
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot…
gpt-5-4-prompting
Internal guidance for composing Codex and GPT-5.4 prompts for coding, review, diagnosis, and research tasks inside the Codex Claude Code plugin.
ideogram4
Prompting patterns for Ideogram 4 text-to-image — best-in-class in-image text rendering and exact color/layout control via structured JSON captions. Use when generating images that need legible on-image text (title cards, thumbnails, logos, signage, CTAs), precise brand colors, or controlled spatial layout. Triggers…