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 laurigates/claude-plugins --skill multi-model-delegationgit clone --depth 1 https://github.com/laurigates/claude-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/laurigates/claude-plugins/multi-model-delegation)<a href="https://agentmods.dev/skills/laurigates/claude-plugins/multi-model-delegation"><img src="https://agentmods.dev/badge/skills/laurigates/claude-plugins/multi-model-delegation/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/laurigates/claude-plugins/multi-model-delegation"><img src="https://agentmods.dev/badge/skills/laurigates/claude-plugins/multi-model-delegation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 6 Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
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 | $0.00041 | $0.02834 |
| Opus 5.5 | $0.00016 | $0.01134 |
| Sonnet 5.5 | $0.00008 | $0.00567 |
| Haiku 4.5 | $0.00004 | $0.00283 |
Grade A, and why
multi-model-delegation 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 — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Multi-Model Delegation
Protocol for consulting other models — kimi, glm, gemini, gpt via the PAL
MCP gateway (chat, consensus) — on design and judgment work, and for
acting on what comes back. The core insight, which inverts the naive
approach:
The value is the disagreement, not the union. Two competent models briefed identically converge on the obvious 80% — the part you'd have written anyway. Where they split is a precise pointer at the one decision that is genuinely load-bearing and underdetermined by the prompt. Resolve the split against the codebase — which usually already decided, and which the models structurally cannot see — never by picking the more confident model.
Treat delegated models as idea generators, never as authorities. Taking the majority answer, or the more confident one, launders a coin-flip into a decision that merely looks researched.
You — the orchestrating Claude session — run the whole consult: you dispatch the PAL MCP calls, collect the replies, and do the judgment steps (diff, adjudicate, synthesize) yourself in the main loop.
The Tool Prefix Is Derived, Not Fixed
PAL's tools are reachable as mcp__<server-name>__<tool>, where
<server-name> is the key the server is registered under — not the product
name, and not the binary. That key can live in any registration scope: a
project .mcp.json, a user-scope entry in ~/.claude.json
(claude mcp add -s user …, which has no .mcp.json at all), or a local one.
So read the registration rather than a file — claude mcp list is
authoritative in every scope, and claude mcp get names the scope that owns it:
claude mcp list # "pal-mcp-server: pal-mcp-server - ✔ Connected"
claude mcp get pal-mcp-server # Scope: Project config (shared via .mcp.json)
With the common registration pal-mcp-server, chat is
mcp__pal-mcp-server__chat; a repo that registers the same binary under a
different key gets that key in the prefix instead. Every
mcp__pal-mcp-server__* name below assumes that registration — substitute the
key claude mcp list reports.
What ships with it
4 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.
- 2d ago Changed · +4 lines 3ee401e49486
- 15d ago Changed · +1 lines 48b437789bb9
- 18d ago Changed · -41 lines 15b2c060089b
- 1mo ago Changed · -10 lines 640a8bfa0f7b
- 1mo ago First seen · 241 lines · 41 tokens per session scan A 9f53c1d0dd7e
multi-model-delegation is a skill published in the GitHub repository laurigates/claude-plugins (59 stars, last pushed yesterday), licensed MIT. It adds 41 tokens to every session and 2,834 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
extract-ssot
Deduplicate repeated markdown (rule files, skill bodies, ADRs, docs) into one named source of truth and migrate every call site to cite it by exact heading. Use when the same prose, literal, or concept recurs across files: 'DRY this prose', 'extract a shared rule', 'single source of truth for X', a value-bump diff…
feature-branch-pr-writing
Use this skill when the user asks to write, draft, create, or improve a PR description for a Shopware core repository PR — AND that PR targets a non-trunk feature branch (not trunk itself). Trigger phrases like "write a PR description", "draft the PR", "what should I put in the PR body". The skill detects the target…
motion-design
Craft rules for anything that moves on a timeline, in video or UI: animate this, motion graphics, title card, intro or outro, transition, explainer that feels like slides, easing, timing, how long to hold text. Gives numeric defaults, an accessibility gate and the tells of generated-looking motion.
performance
Profiling-driven performance optimization: caching, query tuning, N+1 fixes, memory, and load testing. Use to diagnose a bottleneck or set performance budgets.
repo-activity
Scan all git repositories under /repos and report recent activity — last commit age, branch, uncommitted changes — in age-bucketed tables. Use when the user asks for a repo activity overview, "what have I been working on", portfolio status, or which repos are active/dormant.
oslog
Read, stream, and analyze Apple unified logs (OSLog) for iOS/macOS apps. Auto-detects subsystem from Logger usage or bundle identifier. Supports live log show, real-time streaming, and .logarchive analysis with full predicate reference. Use when the user asks to check app logs, stream console output, debug with OSLog…