Chat Agents, Coding Agents, and Workflows
A chat agent helps explore, explain, plan, and review. A coding agent works directly with a codebase and executes implementation tasks. A workflow connects several specialised steps so that ideation, specification, implementation, and testing remain separate.
What You Will Be Able to Decide
- Separate thinking tasks from implementation tasks.
- Choose between chat agents, coding agents, and vibe-coding environments.
- Review agent skills and permissions before use.
- Build a controlled specification, implementation, and testing workflow.
A founder can describe an idea to ChatGPT, Claude, Codex, Claude Code, Lovable, Replit, or another AI product.
These tools may all respond with software-related output. They do not perform the same role.
The first decision is not which brand is best.
The first decision is what type of work is being assigned.
Chat Agents
A chat agent works through a conversation.
It is useful for:
- exploring an idea
- identifying assumptions
- comparing options
- writing requirements
- explaining technical terms
- reviewing a proposal
- improving a prompt
- challenging scope
Think of a chat agent as a consultant working at a whiteboard.
It can help decide what should be built. It may not have direct access to the repository, terminal, runtime errors, or deployment environment.
ChatGPT and Claude can act as chat agents when used primarily through conversation.
Coding Agents
A coding agent works closer to the implementation environment.
It may be able to:
- inspect repository files
- edit code
- run commands
- read errors
- write tests
- update dependencies
- create commits
- follow repository instructions
Codex and Claude Code are examples of tools commonly used in this role.
A coding agent is closer to a developer working inside the project than a consultant discussing the project from outside.
The distinction is important because a planning conversation and an implementation task require different context.
Vibe-coding Environments
Vibe-coding environments such as Lovable and Replit reduce the distance between describing an application and seeing a working interface.
They can be useful for:
- prototypes
- early MVPs
- interface exploration
- internal tools
- validating a workflow
- demonstrating an idea
Technical term
Vibe coding
The term “vibe coding” generally describes building software through natural-language direction while allowing the AI system to make many implementation choices.
The correct response is not to reject vibe coding.
The correct response is to define what must be inspected before the result is trusted.
Technical term
Skill
A skill is a reusable set of instructions, knowledge, templates, or procedures that helps an agent perform a specialised task.
Think of a skill as an operating manual supplied to a capable worker. The worker may understand many general tasks. The manual explains how this organisation expects a particular task to be completed.
A skill may teach an agent how to:
- review architecture
- write a specific type of document
- follow a design system
- deploy to a platform
- inspect security risks
- use a particular tool
A skill should be reviewed before use. Check:
- who created it
- what permissions it assumes
- which tools it invokes
- whether its recommendations fit the current project
- whether it changes files or infrastructure
- whether it contains outdated conventions
Technical term
Agent
An agent is an AI system given a goal, context, tools, and some ability to act.
Agents and Agent Systems
Agent systems such as OpenClaw, Hermes, and Strands may coordinate:
- models
- tools
- memory
- tasks
- schedules
- approvals
- external services
An agent can be more autonomous than a normal conversation. That autonomy increases the importance of permissions and review.
A useful rule is:
Increase autonomy only after the task, limits, and recovery process are clear.
Why the Workflow Matters
A weak workflow asks one tool to:
- understand the business
- choose the scope
- select the architecture
- design the interface
- write the code
- test the result
- declare the work complete
This combines several judgement-heavy tasks into one instruction.
A stronger workflow separates them.
The founder and chat agent clarify the product.
The founder reviews the decision using the knowledge available.
The coding agent implements a bounded task.
Testing produces evidence.
The next prompt responds to that evidence.
This creates a controlled feedback loop.
Questions to Ask a Consultant
- Is this a thinking task or an implementation task?
- Does the tool have access to the repository?
- What files or systems can it change?
- Which decisions should remain with the founder or consultant?
- What evidence will prove that the task is complete?
- How will failed changes be reversed?
- What should the agent do when the requirement is ambiguous?
Key takeaway
Key Takeaway
Use chat agents to improve the decision. Use coding agents to execute a bounded implementation. Use workflows to keep specification, implementation, and testing connected without treating them as the same task.