Capabilities for coding agents

Keep an agent read-only until a human approves

Coding agents can do real data-engineering work, but can also ship a schema change or drop a table.

What makes an agent — model, tools, loop, guardrails

Build the map before touching anything: an agent is a model that reasons, tools that act, a loop that iterates, and guardrails that bound it.

Wire the core MCP servers for a data repo

Wire the three reference MCP servers a data repo needs: filesystem for the working directory, git for history and commits, and GitHub for issues and pull requests.

Audit an MCP server's tool contracts

• Read a tool schema and predict its side effects
• Classify tools by blast radius

Hooks — deterministic gates around an agent

• Explain where hooks run — on tool-call and pre-run events, outside the model
• Match each safety hook to the failure it prevents

Inside a skill — SKILL.md, references, examples, the loading protocol

A skill is the unit of reusable agent behavior: one SKILL.md with front-matter, plus directories the agent loads on demand.

Onboard an agent with the DE Data Toolkit

Onboard an agent onto a data repo with the DE Data Toolkit plugin: the SQL/Spark review skills, the destructive-SQL guard, a secret scanner, and the file/git MCP servers already wired.

Capstone — the guarded agent checklist

Run the whole series against one agent setup: skills that load on demand, tools scoped at the connection, least-privilege credentials, audited tool contracts, hooks that fail closed, and a human gate on anything irreversible.

Production gotchas — untrusted content, destructive actions, permission scope

An agent executes with your credentials, so its context and config are an attack surface.

When an agent does the wrong thing — contain, audit, postmortem

• Contain the damage before diagnosing — rollback first
• Reconstruct what the agent did from the audit trail

MCP Tool Contract Auditor

• Read a tool schema and predict its side effects
• Classify tools by blast radius

Fetch MCP Server

Retrieves a URL and converts it to markdown for the model.

Filesystem MCP Server

Reference server that exposes a sandboxed set of directories as readable and (optionally) writable tools.

Git MCP Server

Exposes local Git operations — status, diff, log, commit, branch — as tools, so an agent can inspect history and prepare changes without shelling out blindly.

GitHub MCP Server

First-party GitHub server exposing repositories, issues, pull requests and code search, plus optional write operations.

DE Data Toolkit

The opinionated starter plugin: SQL/Spark review skills, the destructive-SQL guard, a secret scanner and the file/git MCP servers, wired for data engineers.

Agent-in-the-Loop Review

Coding agents can do real data-engineering work, but can also ship a schema change or drop a table.

Agent Skills (open format)

The Agent Skills format specification — a SKILL.md file with front-matter (name, description, version), plus references/, scripts/, and examples/ directories that an agent loads on demand. The format this platform's Skill object is modelled on — portable across compatible agents. Covers the SKILL.md schema, the directory layout convention, and the agent loading protocol.

Daily Life as a Jenkins Administrator

Jenkins Kubernetes setup As a team, we introduced Jenkins Kubernetes for every user a few months ago.

GSoC 2026 Final Blog: How It Was to Be a Contributor at Jenkins

Hello everyone!

Jenkins and AI Agents: The Trusted Engine for Faster Software Delivery

The 2026 Reality Check By 2026, software development has crossed a threshold.

kitfunso/hippo-memory

Stop re-teaching your agent.

MCP Registry (official)

The official MCP Registry — community-driven registry of MCP servers with a metadata schema for discovery. The canonical place to discover MCP servers — treat entries as *discovery*, never as a trust signal. The metadata schema defines server capabilities, transport, and authentication; essential for building MCP-aware tooling.

Model Context Protocol Specification

definitive reference for transports (stdio, SSE, HTTP), tool definitions, resource templates, prompts, and the security model. The security section is required reading before connecting any MCP server to real data; the transport section explains stdio vs HTTP tradeoffs. The canonical reference for all MCP server implementations on this platform.

Monitoring MWAA-orchestrated ETL pipelines with Amazon OpenSearch Service

Troubleshooting a multi-service, MWAA-orchestrated ETL pipeline often means hunting across scattered Amazon CloudWatch log groups.

oaslananka/kicad-mcp-pro

MCP server for KiCad, connecting AI agents to schematic and PCB workflows, ERC/DRC validation, DFM checks, BOM analysis, and manufacturing preparation—with controlled edits and human-gated release wor