Pydantic Deep Agents is a self-hosted terminal AI assistant and Python framework for building coding, research, and other AI agents. It gives agents tools such as file access, shell commands, planning, memory, sub-agents, sandboxed execution, and MCP connections, and supports different models.
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 vstorm-co/pydantic-deepagents --skill skill-creatorgit clone --depth 1 https://github.com/vstorm-co/pydantic-deepagentsWrote 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/vstorm-co/pydantic-deepagents/skill-creator)<a href="https://agentmods.dev/skills/vstorm-co/pydantic-deepagents/skill-creator"><img src="https://agentmods.dev/badge/skills/vstorm-co/pydantic-deepagents/skill-creator.svg" alt="Measured on agentmods" 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 | $0.00011 | $0.00311 |
| Opus 5 | $0.00005 | $0.00156 |
| Sonnet 5 | $0.00002 | $0.00062 |
| Haiku 4.5 | $0.00001 | $0.00031 |
Grade A, and why
skill-creator 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 8d 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.
What it actually says
Skill Creator
Create a new skill by generating a SKILL.md file with YAML frontmatter.
Steps
- Identify the task pattern the user wants to capture as a skill
- Write clear, specific instructions that an AI agent can follow
- Create a directory with this structure:
my-skill/
├── SKILL.md # Instructions + frontmatter
├── template.md # (optional) Template files
└── example.py # (optional) Example scripts
- The SKILL.md must have this format:
---
name: skill-name
description: "Brief description of what this skill does"
tags: [category1, category2]
version: "1.0.0"
---
# Skill Name
## Purpose
What this skill accomplishes.
## Instructions
Step-by-step instructions for the agent.
## Examples
Concrete examples of input/output.
Guidelines
- Name: lowercase with hyphens (e.g.,
api-client-generator) - Description: one sentence, starts with a verb
- Instructions: specific enough that another agent can execute without context
- Include edge cases and error handling guidance
- Add examples showing expected input and output
- Place resources (templates, schemas) as separate files next to SKILL.md
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.
- 8d ago First seen · 55 lines · 11 tokens per session scan A 57e65c236dd3
skill-creator is a skill published in the GitHub repository vstorm-co/pydantic-deepagents (1,059 stars, last pushed 16d ago), licensed MIT. It adds 11 tokens to every session and 311 once invoked, about $0.0001 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
Skillify
Interview the user about a workflow they keep repeating, then generate a reusable SKILL.md and write it to disk so the next session starts with the skill already loaded.
Sandbox Tool Builder
How to build a NEW tool of your own by writing its code in your sandbox, testing it, and forging it into a live, callable tool with env(action="forgetool"). Use when no existing tool does what you need and you can implement it as a small script. Multiple tools + skills can be saved together as one reusable Sandbox…
Self-Modifying Environment
How to inspect and edit your OWN operating environment at runtime — system prompt, active tools, and skills — using the env tool. Use when you need to change your own capabilities (turn a tool on/off, rewrite your instructions, author a new skill) or save them for this session.
Get Unstuck
Pause, surface the current plan + open questions, and suggest the next concrete step so the agent (or operator) can recover from a deadlock or analysis paralysis.
torchforge-rl-training
Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.
pydantic-ai
Build production-ready AI agents with PydanticAI — type-safe tool use, structured outputs, dependency injection, and multi-model support.