AI Developer Tooling
Use coding assistants, MCP servers, and agent harnesses safely.
Understand how a coding assistant gathers context and calls tools, build a small Model Context Protocol server with validated inputs, and design an agent harness for a team codebase with permissions, sandboxing, evaluation, and budgets.
3 questions
Questions
Easy
1.5 h
Explain how a coding assistant assembles its context, which component decides whether each proposed action runs, and why an instruction found inside a file cannot grant itself permission.
Medium
3 h
Implement the request handler of a small read-only MCP server that validates tool arguments from their declared schemas and puts each failure in the right channel, then review the trust model of adding a shell tool and installing a third-party server.
Hard
4 h
Design one agent harness for a fictional payroll company's developer machines and its CI agent, covering permissions, sandboxing, extensions, hooks, evaluation, and budgets. Then decide from predeclared rules which configuration rolls out first, where, and with which permissions.
What you will be able to do
- Explain the agent loop and which steps the model and the harness each control
- Name the sources of an assistant's context and when each one loads, including instruction files loaded at session start and on demand
- Apply allow, ask, and deny rules in order to decide whether a proposed action runs, asks, or is blocked
- Explain why an instruction inside a file or tool result gains no authority, and which layers still hold when the model follows it
- Describe what installing an MCP server or a skill adds to what an agent can do and see
- Explain what compaction removes and where a constraint must live to survive it
- Explain the host, client, and server roles and what the stateless 2026-07-28 revision puts in every request
- Implement a request handler that checks the JSON-RPC envelope, per-request `_meta`, and protocol version, with the defined error codes
- Validate tool arguments from the declared input schema and return each failure as a tool result the model can act on
Hands-on exercises
Interview rounds that use this topic
Topic boundaries
This topic covers the tools an engineer uses in their own AI-assisted development: coding assistants, MCP servers, skills, hooks, repository instructions, and agent harnesses. Tool Calling & Agents covers the tools a model calls inside your application. AI for Team and Tech Leads covers whether a team adopts these tools, the rules staff follow, and measuring the effect on the team's work.
Related on CodeSnatch
System Design
Teaches scoped, expiring credentials and allow or deny decisions, the basis for giving each MCP server and CI agent a narrow token.
Practice problem
Code the side that decides whether a model-proposed tool call runs, taking identity only from trusted context.
AI Engineering
Covers tool description changes, supply chain review, and separating untrusted reading from privileged execution in an agent.
System Design
Its secure development section covers secrets managers, least privilege, egress filtering, and pinning packages by hash.
AI Engineering
Picks up where this topic stops: judging from delivery data whether coding assistants actually helped a team.
Behavioral Interview
Frames stories about choosing between tools or configurations with explicit criteria, as harness decisions require.
Community article
A practitioner's account of tasks where assistants save time, where they cost time, and the review habits that followed.
Community snippet
Redacts secrets by key name and by pattern before logging, a habit that applies to agent transcripts and audit logs.