mozaiks: Skill for Claude Code

.agents/skills/agentgenerator-change/SKILL.md

agentgenerator-change is a skill for Claude Code from BlocUnited-LLC/mozaiks. It costs 33 tokens per session (1,568 once invoked), scanned A, original, MIT.

A guide for changing the AgentGenerator, the part of the system that creates AI agents and workflow bundles. It covers prompts, structured outputs, generated scaffolds, shared prompt instructions, tools, and safety guidance.

In plain words
What is it for?
Use it when changing AgentGenerator files, agent or workflow prompts, generated workflow structures, tool or context injection, safety rules, or related tests.
Why use it?
It helps keep generated agents and workflows consistent with the project's contracts and safety rules when the generator changes.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: installed under .agents/ (shared by several agents).

This is BlocUnited-LLC/mozaiks's own configuration. It tells Claude Code how to work on mozaiks itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything mozaiks configures →

Reuse

Borrowing it

Nothing to install: this file belongs to BlocUnited-LLC/mozaiks. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/BlocUnited-LLC/mozaiks/main/.agents/skills/agentgenerator-change/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/BlocUnited-LLC/mozaiks

Made for: Claude Code.

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 agentgenerator-change

README.md
[![agentmods](https://agentmods.dev/badge/skills/blocunited-llc/mozaiks/agentgenerator-change/github.svg)](https://agentmods.dev/skills/blocunited-llc/mozaiks/agentgenerator-change)
Your own site
<a href="https://agentmods.dev/skills/blocunited-llc/mozaiks/agentgenerator-change"><img src="https://agentmods.dev/badge/skills/blocunited-llc/mozaiks/agentgenerator-change/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 agentgenerator-change

Your own site · 80×15
<a href="https://agentmods.dev/skills/blocunited-llc/mozaiks/agentgenerator-change"><img src="https://agentmods.dev/badge/skills/blocunited-llc/mozaiks/agentgenerator-change.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 33 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,568 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00033 $0.01568
Opus 5 $0.00016 $0.00784
Sonnet 5 $0.00007 $0.00314
Haiku 4.5 $0.00003 $0.00157

Measured 9d ago against content hash 4c767c6606e1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

agentgenerator-change 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 9d 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.

.agents/skills/agentgenerator-change/SKILL.md · 143 lines

How it starts

The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Before starting: git fetch origin && gh pr list --state open && git log origin/main --oneline -3 — if another agent has an open PR touching the same files you need, wait for it to merge or branch off it instead of main.

Use this skill when a change touches AgentGenerator specifically.

Typical triggers:

  • factory_app/workflows/AgentGenerator/**
  • AgentGenerator workflow prompts
  • agent or workflow bundle structured outputs
  • generated workflow scaffolds
  • generated agent prompts
  • generated transition_graph.yaml, tools.yaml, structured_outputs.yaml, or context_variables.yaml
  • universal prompt injection
  • MCP or tool capability injection
  • workflow or agent safety rules
  • agent-generated runtime behavior guidance
  • AgentGenerator tests or fixtures

Inspect first:

  • factory_app/workflows/AgentGenerator/agents.yaml
  • factory_app/workflows/AgentGenerator/structured_outputs.yaml
  • factory_app/workflows/AgentGenerator/middleware.yaml
  • factory_app/workflows/AgentGenerator/transition_graph.yaml
  • factory_app/workflows/AgentGenerator/tools.yaml
  • factory_app/workflows/AgentGenerator/context_variables.yaml
  • factory_app/workflows/AgentGenerator/tools/hook_universal_prompts.py
  • capability-hook files if they exist in this workflow later, such as hook_mcp_capabilities.py or agent_capabilities.py
  • factory_app/workflows/AgentGenerator/tools/tool_planning.py
  • factory_app/workflows/AgentGenerator/tools/workflow_converter.py
  • factory_app/workflows/AgentGenerator/tools/generate_and_download.py
  • docs/architecture/workflows/workflow-authoring-contracts.md
  • docs/architecture/builder/agentgenerator-output-assembly-contract.md
  • docs/architecture/workflows/workflow-routing-transitions.md
  • the narrowest matching AgentGenerator tests before editing:
    • tests/test_agentgenerator_workflow_converter.py
    • tests/test_agentgenerator_tool_planning.py
    • tests/test_agentgenerator_ui_quality_gate.py
    • tests/test_agentgenerator_generate_and_download_collection.py
    • tests/test_agentgenerator_generate_and_download_persistence.py

Core truth:

  • AgentGenerator is one workflow inside the broader factory build workflow_sequence.
  • It generates workflow and agent bundles, not app modules or persistent app pages.
  • AppGenerator generates app bundle artifacts; AgentGenerator generates AI workflow artifacts.
  • It should not own runtime substrate behavior.
  • It should not own app page or module generation.
  • It should not inject hosted or private product assumptions into generated agents.
  • It should produce workflows that respect the current workflow authoring contracts.
  • Generated workflow tools, hooks, and helpers stay workflow-local unless the current shared factory infrastructure explicitly owns the shared helper.

Boundary rules:

  • Do not treat AgentGenerator as the whole build system.
  • Do not change workflow_sequence composition from this skill; use factory-build-workflow-change when the change widens into extension_registry.json, sequence design, transitions, entrypoints, or cross-workflow ownership.
  • Do not generate app modules or persistent app pages; that belongs to AppGenerator.
  • Do not generate runtime transport, platform, or substrate code.
  • Do not inject private hosted-product names, assumptions, or proprietary examples into prompts, scaffolds, tests, or docs.
  • Do not use stale workflow fields like startup_mode when the current contract uses workflow_startup_mode.
  • Do not create fake MCP or tool declarations unsupported by runtime or current workflow contracts.
  • Do not bypass workflow-local handoff semantics.
  • Do not use workflow_sequence as a HITL substitute.

Common change types:

  1. Prompt changes:
    • inspect agents.yaml, hook_universal_prompts.py, and the nearest prompt-hygiene tests together
    • preserve universal compliance and lane-discipline guidance
  2. Structured output changes:
    • inspect structured_outputs.yaml plus the workflow authoring contract docs and converter tests
    • keep field names and enums aligned with current loader expectations
  3. Universal prompt hook changes:
    • inspect hook_universal_prompts.py, middleware.yaml, and tests that validate prompt or UI-safety guidance
    • keep universal behavior separate from file-generation or pattern-specific hooks
  4. MCP or tool capability hook changes:
    • inspect tool_planning.py, tools.yaml, and any capability hook files if present
    • add only runtime-supported tool declarations and workflow UI requirements
  5. Workflow scaffold or file contract changes:
    • inspect workflow_converter.py, generate_and_download.py, and workflow authoring contract docs together
    • keep generated bundle files workflow-local and contract-bound
  6. Handoff generation changes:
    • inspect transition_graph.yaml, converter tests, and authoring docs together
    • preserve workflow-local handoff semantics and exact cross-reference names
  7. Tool generation changes:
    • inspect tools.yaml, tool_planning.py, and workflow UI contract tests together
    • keep tool declarations aligned with supported UI and lifecycle surfaces
  8. Workflow safety or human-review guidance:
    • inspect prompt injections, workflow authoring docs, and quality-gate tests together
    • keep HITL semantics explicit inside the workflow bundle, not in workflow_sequence
  9. Agent role or persona generation:
    • inspect agents.yaml, structured_outputs.yaml, and downstream scaffold expectations together
    • keep generated agents scoped to workflow behavior, not runtime ownership
  10. Test or fixture updates:
  • update the narrowest workflow-bundle fixture that changed
  • keep examples provider-neutral and OSS-safe

Read the full file on GitHub · 143 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. 9d ago First seen · 143 lines · 33 tokens per session scan A 4c767c6606e1

Subscribe to this mod's changes

agentgenerator-change is a skill published in the GitHub repository BlocUnited-LLC/mozaiks (25 stars, last pushed today), licensed MIT. It adds 33 tokens to every session and 1,568 once invoked, about $0.0002 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.

Related

Other skills, from other repositories

building-pydantic-ai-agents

Build AI agents with Pydantic AI — tools, capabilities (including on-demand loading), structured output, streaming, testing, and multi-agent patterns. Use when the user mentions Pydantic AI, imports pydanticai, or asks to build an AI agent, add tools/capabilities, defer capability loading, stream output, define agents…

pydantic/pydantic-ai · 85 tokens

migrating-langchain-to-pydantic-ai

Migrate Python LangChain or LangGraph applications to Pydantic AI. Use for LangChain agents, chains, LCEL, or direct LangGraph graphs, persistence, interrupts, and streaming. Do not use for migrations centered on createdeepagent or Deep Agents harness features.

pydantic/pydantic-ai · 69 tokens

agentfield-use

Whenever you have a discrete task to perform — one the user delegated, or one that arose inside your own work — check FIRST whether an installed AgentField agent covers it, and offload to it by default when one does. Coverage, not task size, is the test: even a small job goes to a covering agent. The check is cheap …

Agent-Field/agentfield · 259 tokens

agentfield

Design and ship a multi-agent system on AgentField. Use when the user asks to build, scaffold, design, or run an agent, reasoner network, multi-agent backend, or 'an agent that does X' — whenever the work would otherwise be a single LLM call or a flat LangChain/CrewAI/AutoGen chain. The skill produces composite…

Agent-Field/agentfield · 127 tokens

agentfield-personal

Build and install a personal AI agent on this machine's AgentField: real source in /agentfield-agents, packaged with agentfield-package.yaml, installed with af install, started with af run, registered on the local control plane, and visible in AgentField Desktop with a keys form and an auto-start toggle. Use when the…

Agent-Field/agentfield · 134 tokens

hephaestus-network

Use when the user types $hephaestus-network, /hep-network, or /agentlas-network, mentions @Hephaestus, or asks Agentlas to staff a durable goal from registered Local, owner Cloud, and public Hub agents or teams. The active host LLM staffs each turn; the exact roster remains goal-bound until explicit completion.

agentlas-ai/Agentlas-OS · 75 tokens