Prepared Prompts
Reusable prompts with fill-in inputs like company name, date, goal, or output style.
No-code and low-code Agent Kit builder
Agent Kits bundle prompts, skills, guardrails, templates, examples, and context so your AI can perform specific jobs consistently. AgentKitForge helps you build, run, import, package, and install them from one desktop app.
Run kits with OpenAI, Claude, Gemini, Ollama, or custom compatible providers, then copy, download, package, or install the result without leaving Forge.
Create a client-ready research brief from notes, source links, and a target audience.
Research brief.docx
Start here
An Agent Kit is a reusable package for a specific AI workflow. Instead of rebuilding the same instructions every time, you save the workflow once and run it again whenever you need it.
Reusable prompts with fill-in inputs like company name, date, goal, or output style.
Step-by-step instructions for how the AI should complete part of the job.
Guardrails for what the AI should avoid, require, or escalate.
Output formats the AI should follow, such as a report, memo, checklist, or summary.
Sample inputs and outputs that show what good results look like.
Reference material, notes, or examples the AI can use while working.
When you run a kit, AgentKitForge collects the inputs, prepares the prompt, sends it to your selected AI provider, and lets you copy or download the result.
Read the beginner guideA prepared prompt for reviewing a monthly workbook, inputs for company name and reporting period, skills for formula review and trend analysis, policies requiring human review, and a client-ready summary template.
Skills for investigating pod failures, service outages, and logs; read-only policies by default; prepared prompts for incident triage; and templates for incident summaries.
A prepared prompt that asks for client name, topic, and tone, plus a memo template, style examples, and policies for clarity and human review.
One app
No need to jump between tools
Use your provider of choice
Build with AI or guided forms
Required inputs are collected before running
Validate before use
Generate outputs inside Forge
Download document-like outputs
Package .agentkit.zip
Export one-file Markdown
Install to Codex / Claude Code
Build your way
Describe the kit you want, iterate on draft sessions, request changes, and save when ready.
Use structured forms for purpose, inputs, prepared prompts, outputs, and validation.
Refine imported or saved kits with Edit with AI or the Guided Editor.
Start from templates, drafts, or advanced draft JSON when a repeatable shape already exists.
Use inside Forge
Providers and installs
Run Agent Kits inside Forge with OpenAI, Anthropic Claude, Google Gemini, Ollama, or a custom OpenAI-compatible endpoint.
Install Agent Kits into supported agent tools like Codex and Claude Code.
Prepare one-file Markdown bundles for web assistants like ChatGPT and Claude.
Package .agentkit.zip, export one-file Markdown, import from zip/folder/Git,
and keep portable versions.
Package and share
Package .agentkit.zip files, export one-file Markdown, import from
.agentkit.zip, local folders, or Git repositories, and keep shareable kit
artifacts without leaving the app.
Install on Local Agent
Install Agent Kits into Codex and Claude Code when a kit supports those local agent formats. This is separate from AI provider setup and does not imply every model backend supports persistent skills.
CLI for developers
The desktop app is product-first, but the same core supports CLI workflows for teams that want scripts, checks, and repeatable exports.
View CLIagentkit.yaml
AGENTKIT.md
skills/research/SKILL.md
workflows/review.md
templates/brief.md
exports/agentkit.md Preview
Forge builds kits. Market will distribute and store kits through a public marketplace and, later, private organization repositories. Kits can be built and used with many AI providers; Market distribution will stay independent of provider choice.
Read the market preview