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 DDS-Solutions/AI-TadPole-OS --skill gemma4-localgit clone --depth 1 https://github.com/DDS-Solutions/AI-TadPole-OSWrote 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/dds-solutions/ai-tadpole-os/gemma4-local)<a href="https://agentmods.dev/skills/dds-solutions/ai-tadpole-os/gemma4-local"><img src="https://agentmods.dev/badge/skills/dds-solutions/ai-tadpole-os/gemma4-local/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/dds-solutions/ai-tadpole-os/gemma4-local"><img src="https://agentmods.dev/badge/skills/dds-solutions/ai-tadpole-os/gemma4-local.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00036 | $0.01009 |
| Opus 5.5 | $0.00014 | $0.00404 |
| Sonnet 5.5 | $0.00007 | $0.00202 |
| Haiku 4.5 | $0.00004 | $0.00101 |
Grade A, and why
gemma4-local 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 12d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
[!IMPORTANT] AI Context & Knowledge Heritage
- Subsystem: Agent Skills Registry / gemma4-local
- Architecture:
@docs ARCHITECTURE:Documentation- Failure Path: Token mismatch, raw thought leakage, or invalid tool string delimiters.
- Observability: Traceability via
execution/parity_guard.py([SKILL:GEMMA4])
Gemma 4 (gemma4) Local Model Interaction Protocol
Target Model Identifier: Explicitly formatted as
gemma4(no spaces/hyphens). Purpose: Enforce exact control token delimitation, thought channel containment, and tool parameter serialization for local execution.
🎯 Progressive Disclosure & L3 Reference Index
Read REQUIRED core token rules below; consult deep L3 Reference Guides on demand:
| L3 Reference Guide / Tool | Purpose | Trigger / When to Load |
|---|---|---|
references/prompt_templates_and_examples.md |
Multi-turn dialogue transcripts, raw tool JSON syntax, edge-case delimiters | Custom local server templates & raw token debugging |
scripts/gemma_prompt_linter.py |
Deterministic prompt token validator with string delimiter checks and --test suite |
Automated validation of prompts dispatched to local gemma4 runtime |
1. Dialogue & Reasoning Control Tokens
| Control Token | Purpose | Positioning / Scope |
|---|---|---|
| `< | turn>system...<turn |
>` |
| `< | think | >` |
| `< | channel>thought...<channel |
>` |
| `< | turn>user/< |
turn>model` |
2. Tool & Function Calling Tokens
| Token Pair | Function |
|---|---|
| `< | tool>...<tool |
| `< | tool_call>...<tool_call |
| `< | tool_response>...<tool_response |
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.
- 12d ago Changed · +35 lines c39a5078f8da
- 1mo ago First seen · 63 lines · 36 tokens per session scan A f58e1b26872e
gemma4-local is a skill published in the GitHub repository DDS-Solutions/AI-TadPole-OS (8 stars, last pushed today), licensed MIT. It adds 36 tokens to every session and 1,009 once invoked, about $0.0001 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 2026-09-03.
Other skills, from other repositories
prompt-cache-optimizer
Audit and optimize Pisper system-prompt and tool-schema token overhead while preserving stable prompt-cache prefixes, permissions, and runtime behavior. Invoke only for explicit prompt or tool-context optimization work.
developing-genkit-js
Develop AI-powered applications using Genkit in Node.js/TypeScript. Use when the user asks about Genkit, AI agents, flows, or tools in JavaScript/TypeScript, or when encountering Genkit errors, validation issues, type errors, or API problems.
v4-best-practices
Use when working with deepseek-v4-pro or deepseek-v4-flash in thinking mode on multi-step or plan-driven tasks. Provides rules to prevent stale references, unverified plan assumptions, and vague plan output.
ax-gen
This skill helps an LLM generate correct AxGen code using @ax-llm/ax. Use when the user asks about ax(), AxGen, generators, forward(), streamingForward(), validation, assertions, streaming assertions, field processors, step hooks, self-tuning, or structured outputs. For MCP clients, transports, prompts, resources…
ax-signature
This skill helps an LLM generate correct DSPy signature code using @ax-llm/ax. Use when the user asks about signatures, s(), f(), field types, string syntax, fluent builder API, validation constraints, or type-safe inputs/outputs.
ax-rust-signature
Use when writing Rust code with axllm for string signatures, field descriptors, JSON schema output, validation, and typed tool argument shapes.