Callee

Callee is a self-reported CLI tool built in Go, designed to enable users to define AI agent workflows using Markdown or YAML within a repository. The author describes it as a deterministic workflow engine for composing pipelines and graph-style workflows from subagents across different coding hosts (e.g., Codex, Claude, Copilot). It operates as a single-binary runtime with host integrations and is positioned as a lightweight, local-first system.

The description states that Callee supports repo-defined agents, simple composition, Markdown-first authoring, and local execution. It does not claim to have any revenue, customers, or traction beyond the author's own use and development.

The description states that Callee is a CLI tool written in Go, intended for defining and executing AI agent workflows through Markdown or YAML. It supports:

• Composition of pipelines, loops, and graph-style workflows.
• Execution via providers like Codex, Claude, Copilot, etc.
• Repository-native definitions without heavy configuration.
• Host integrations to support multiple coding environments.
• Local execution with a single-binary runtime.

It builds on prior tools such as Norma Runtime, go-adk-acpagent, and others, but introduces a new composition model and authoring experience focused on simplicity and local-first design.

Inference: The product is not a SaaS offering or cloud-hosted solution. It is a command-line utility for developers to define AI workflows locally.

The description states that Callee was inspired by the desire to build AI pipelines, loops, and graph-style workflows out of subagents from different coding hosts.

It positions itself as:

• A simple, repo-defined system.
• Supporting deterministic workflows.
• Enabling cross-host agent composition.
• Focused on local execution and configuration-light design.

The author emphasizes a KISS (Keep It Simple, Stupid) approach, resisting complexity in favor of usability and portability.

Inference: Callee is positioned as a lightweight developer tool for composing AI workflows locally, not a full-fledged orchestration platform or enterprise solution.

The description does not name specific customers or personas. However, it implies:

• Developers who work with AI agents and coding hosts.
• Users looking to define and run AI workflows in a local-first, repo-native way.
• Those interested in cross-host composition of subagents.

It is implied that the tool targets technical users who are comfortable with CLI tools, Markdown/YAML, and local development environments.

Inference: The ICP appears to be technical developers or engineers working with AI agents and coding environments, but no explicit segmentation or customer data is provided.

The description does not state anything about pricing, monetization, or business model. It is a self-reported tool built by one person (Alexey Samoylov) for personal use and hackathon submission.

Inference: There is no evidence of a commercial business model, revenue, or pricing structure.

The description states:

• Built in Go, as a single-binary CLI runtime.
• Definitions are stored in Markdown or YAML.
• Supports ACP-compatible providers (e.g., Codex).
• Integrates with host plugins for different coding environments.
• Uses existing tools like Norma Runtime, go-adk-acpagent, and others as a foundation.
• Designed to be local-first, configuration-light, and simple.

It also mentions that the author avoided complexity by keeping the system repo-native, single-binary, and KISS.

Inference: The tool is technically minimalistic, built for developers who value simplicity and local execution. It does not appear to be a cloud-based or hosted solution.

The description states that this project was submitted to the OpenAI 2026 hackathon, and that it is a personal project by one developer (Alexey Samoylov). No evidence of:

• Revenue
• Customers
• Adoption
• Usage metrics
• Product-market fit
• Community or user feedback

The author notes accomplishments such as ease-of-use and host integration, but these are self-reported.

Inference: There is no traction or maturity evidence beyond the author’s own development and submission to a hackathon.

The description does not mention any competitors. It focuses on Callee’s own design principles and features, such as:

• Repo-defined workflows
• Cross-host agent composition
• Local execution
• Markdown-first authoring

It is implied that the tool competes with or complements existing AI agent orchestration tools, but no direct comparison or competitive analysis is provided.

Inference: No known competitors are mentioned. The tool appears to be in a niche space of local-first AI workflow definition, possibly overlapping with developer tooling or AI agent composition platforms.

• No traction or adoption evidence: The tool is described as a personal project, not a product with users.
• Self-reported claims only: All features and benefits are stated by the author without external validation.
• Single-person development: No team or organizational support is evident.
• Unclear commercial viability: No pricing, monetization, or business model is described.
• Limited scope: The tool is focused on local execution and simple workflows; it may not scale to enterprise or complex use cases.

Inference: The project is in a very early stage with no commercial or user validation. It may be a prototype or proof-of-concept rather than a product ready for market.

1. What specific workflows or tasks are you using Callee to automate today?
2. How do you plan to scale beyond a single developer’s use case?
3. Are there any users or early adopters of Callee outside of your own development?
4. What is the long-term vision for monetization or commercialization?
5. How does Callee differ from other tools in this space, and what are its unique advantages?
6. What are the technical limitations or constraints of the current architecture?