prompt-oss-specialist

prompt-oss-specialist is an agent for Claude Code from frankxai/agentic-creator-os. It costs 152 tokens per session (1,076 once invoked), scanned A, original, Apache-2.0.

A formatter for prompts used with open-source language models such as Llama, Mistral, Qwen, Yi, and DeepSeek-R1. It applies each model family's required conversation format, including its special control markers.

In plain words
What is it for?
Use it when adapting a prompt to a particular open-source model or fine-tune. It covers formats such as ChatML, Mistral instruction blocks, Llama formats, and DeepSeek-R1 reasoning blocks.
Why use it?
Different models expect prompts to be arranged differently, and a hand-written format can silently reduce response quality. It helps avoid sending a model the wrong conversation structure.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

Part of the agentic-creator-os plugin — 133 commands, 68 agents shipped together

Good fit Use it when adapting a prompt to a particular open-source model or fine-tune. It covers formats such as ChatML, Mistral instruction blocks, Llama formats, and DeepSeek-R1 reasoning blocks.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/frankxai/agentic-creator-os/prompt-oss-specialist
Install

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.

Clone the repo
git clone --depth 1 https://github.com/frankxai/agentic-creator-os

Made for: Claude Code.

Or install agentic-creator-os, the plugin that ships this one along with the rest of its 133 commands, 68 agents.

Wrote 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.

agentmods badge for prompt-oss-specialist

README.md
[![agentmods](https://agentmods.dev/badge/agents/frankxai/agentic-creator-os/prompt-oss-specialist/github.svg)](https://agentmods.dev/agents/frankxai/agentic-creator-os/prompt-oss-specialist)
Your own site
<a href="https://agentmods.dev/agents/frankxai/agentic-creator-os/prompt-oss-specialist"><img src="https://agentmods.dev/badge/agents/frankxai/agentic-creator-os/prompt-oss-specialist/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.

agentmods 80×15 button for prompt-oss-specialist

Your own site · 80×15
<a href="https://agentmods.dev/agents/frankxai/agentic-creator-os/prompt-oss-specialist"><img src="https://agentmods.dev/badge/agents/frankxai/agentic-creator-os/prompt-oss-specialist.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 152 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,076 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00152 $0.01076
Opus 5 $0.00076 $0.00538
Sonnet 5 $0.00030 $0.00215
Haiku 4.5 $0.00015 $0.00108

Measured 11d ago against content hash 90c664ca3b24, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

prompt-oss-specialist 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 11d 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.

.claude/agents/prompt-oss-specialist.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prompt OSS Specialist

Mission

Convert any prompt into the correct chat-template format for the target open-source model. Eliminate the #1 silent quality killer: hand-written template strings.

Canonical OSS chat templates

Model family Template Notes
Llama 3 / 3.1 / 3.2 <|begin_of_text|><|start_header_id|>{role}<|end_header_id|>\n\n{content}<|eot_id|> Special tokens. Do NOT hand-write — use tokenizer.apply_chat_template().
Mistral (v0.3+) [INST] {content} [/INST] Has system role v0.3+. Pre-v0.3, embed system inside first INST block.
Qwen / most fine-tunes ChatML: <|im_start|>{role}\n{content}<|im_end|> Cross-model standard for many newer OSS releases.
Yi ChatML Same as Qwen.
DeepSeek-R1 ChatML with reasoning in <think> blocks Reasoning surfaces in the assistant message itself, not API metadata.
CodeLlama, older Llama 2 Llama 2 format: <s>[INST] <<SYS>>\n{system}\n<</SYS>>\n\n{user} [/INST] Legacy.

When to invoke

  • @prompt-conductor dispatches with target lab = oss.
  • "make this work on Llama", "convert for Mistral", "use ChatML", "render with apply_chat_template".
  • Reviewing any system prompt before publish to prompt-library with lane: oss.

Hard rules

  • Never hand-write chat-template strings. Always render via tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True). This is the #1 cause of silent quality loss in OSS deployments.
  • Never assume ChatML works on Llama 3. Llama 3 uses its own special tokens; ChatML on Llama 3 silently degrades.
  • Never strip <think> blocks from DeepSeek-R1 output if you want reasoning visible. They're inline, not metadata.
  • Always check the model card's chat_template field in tokenizer_config.json before assuming a format. Fine-tunes inherit base + change template often.
  • For older Mistral (pre v0.3): no system role exists; embed system content inside the first [INST] block with a delimiter.

Read the full file on GitHub · 74 lines

Changes

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.

  1. 11d ago First seen · 74 lines · 152 tokens per session scan A 90c664ca3b24

Subscribe to this mod's changes

prompt-oss-specialist is an agent published in the GitHub repository frankxai/agentic-creator-os (10 stars, last pushed yesterday), licensed Apache-2.0. It adds 152 tokens to every session and 1,076 once invoked, about $0.0008 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 2026-08-31.

Related

Other agents, from other repositories

prompt-engineer-pm

Owns the AI product's PROMPT discipline — versioning, registry, prompt-as-code, prompt review, prompt-vs-fine-tune decisions. The PM-side architect for everything the product sends to a model. NOT to be confused with query-refiner-pm (which refines USER queries TO great-pm).

VandanaAjayDubey111/great-pm · 71 tokens

llm-integration-agent

LLM entegrasyon görevlerini üstlenir. Model API çağrıları, prompt tasarımı, tool-use şemaları, token/maliyet yönetimi, LLM çıktı doğrulama.

boranesn/agentic-base · 47 tokens

ai-evaluator

Designs and runs AI product evaluation frameworks: error analysis, eval suite design, LLM-as-judge pipelines, human eval protocols, regression testing plans, and improvement flywheels. Use this agent when the user is building an AI-powered feature and needs to define how to measure quality, catch regressions, or…

Productfculty-aipm/PM-Copilot-by-Product-Faculty · 282 tokens

prompt-reviewer

Reviews LLM prompt quality against prompt-master principles. Checks clarity, structure, examples, compression, positive framing. Use after writing or modifying LLM prompts.

stepanenkoviktor0110-boop/ai-dev-methodology-codex · 35 tokens

prompt-engineer

Prompt engineering specialist that creates or refines prompt artifacts using the embedded Prompt Engineering Bible. Use whenever creating or changing system prompts, agent prompts, instruction files, prompt registries, or other behavior-governing prompt assets.

chieflatif/vibeos-plugin · 48 tokens

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens