For builders operationalizing agentic work.

MCP and skills made the agent harness more important than the model choice.

Week 21 of 2026 · May 23, 2026

Big read

W21's read is that agent performance is increasingly shaped by harness quality: what tools the agent can call, which skills it can load, what data it can reach, and what policy gates constrain action. Model quality still matters, but a stronger model inside a weak harness underperforms a capable model with the right connectors, memory, and verifier.

Technique of the week

Build

Harness engineering

Design the environment around the agent: tools, files, skills, permissions, memory, and checks.

Harness engineering is how agent behavior becomes repeatable. It also creates portability: the same workflow can often move across Claude, Codex, Cursor, OpenCode, or other MCP-capable runtimes.

Goal: State the outcome and the boundary of delegated work.
Context: Give the agent the sources, files, examples, and constraints it needs.
Tools: Limit actions to the connectors, commands, and systems required for the job.
Verifier: Define how the output is checked before it is trusted.
Escalation: Name what requires human review, approval, or rollback.

MCP connectors
tool allowlists
skill bundles
workspace permissions

Sources Addy Osmani, Agent Harness Engineering

New agent capabilities

2026-05-23 · OpenAI · Build

Codex

Workspace-bound coding assistance increasingly organized around issues, tests, review, and automation hooks.

The durable read is not any single coding model; it is the emergence of a build harness where context, tools, and verification are part of the workflow.

Sources Cursor MCP and background agent guide

2026-05-23 · Anthropic · Cowork

Claude / Claude Code

Persistent instructions, skills, and agentic coding workflows point toward reusable professional work patterns.

Claude-style workflows are strongest when the human supplies policy, examples, and review criteria that survive beyond one chat.

Sources Claude Skills and MCP Servers practitioner guide

2026-05-23 · Microsoft · Automate

Copilot / Scout

The Microsoft agent surface is moving toward M365-grounded action, background coordination, and enterprise controls.

The enterprise differentiator is governed access to mail, calendar, files, identity, and policy, not generic answer quality.

Sources Cursor MCP and background agent guide

New skills and connectors

2026-05-23 · MCP-capable agents · Connector

Connector-backed workflow

Agents increasingly use connectors to reach source systems instead of relying on pasted context.

Connectors turn chat into work by letting agents read the system of record and return traceable output.

Sources Claude Skills and MCP Servers practitioner guide

2026-05-23 · Claude / Cursor / Codex / Hermes · Skill

Reusable skills

Repeated workflows are being packaged as skills, rules, plugins, and templates.

Skills are the portability layer for operating knowledge; they prevent teams from re-teaching the same workflow every week.

Sources Claude Skills and MCP Servers practitioner guide

Proof of value

Evidence · Confirmed

Practitioner teams · Coding, research, operations, and professional drafting

The strongest reported wins come from bounded workflows with clear checks, not open-ended autonomy.

Treat value claims as credible when the workflow, baseline, and verifier are visible. Treat broad percentage claims without methods as directional at best.

Sources Agent Harness Engineering

Enterprise readiness

Permissioning

Agents need least-privilege access scoped to the workflow, not broad user-equivalent authority by default.

Verification

Every higher-autonomy workflow needs a deterministic check, source trail, rubric review, or human approval gate.

Auditability

Background agents should produce inspectable logs of prompts, tool calls, approvals, outputs, and state changes.

Cost

Loops need budgets and stop conditions because repeated agent calls can turn productivity experiments into runaway spend.

Scorecard

As of 2026-05-23

Mode	Leading pattern	Representative tools	Control gap
Chat	Structured context and critique loops	ChatGPT, Claude, Copilot Chat	Quality still depends on the user's review discipline.
Cowork	Human-supervised delegation with persistent project memory	Claude Cowork, Microsoft Copilot, Cursor	State, approvals, and source grounding must be visible.
Build	Workspace-bound agents with tools, tests, and worktrees	Claude Code, Codex, Cursor, OpenCode	Verification quality determines whether speed becomes rework.
Automate	Scheduled loops with state, tools, and escalation gates	Codex Automations, Microsoft Scout, Hermes, OpenClaw	Always-on agents need identity, audit trails, budgets, and stop conditions.

Try this

Run a Harness engineering experiment

Expected outcome: A reusable workflow contract and a clearer read on whether the task is ready for cowork, build, or automate mode.

Pick one recurring weekly task with a clear definition of done.
Write a one-page loop contract: goal, context, tools, verifier, stop condition, and escalation rule.
Run it manually once with an agent and record where the verifier was weak.
Only automate the task after the verifier catches the most likely failure mode.

Watchlist

Next 7 days

Copilot and Scout agent releases

Microsoft's advantage is governed enterprise context; any new background or M365 action capability changes the automation surface.

Next 7 days

Claude and Codex skill ecosystems

Reusable skills and plugins are the leading indicator that agentic workflows are becoming products, not prompts.

Next 30 days

Hermes, OpenClaw, OpenCode, and adjacent OSS harnesses

Open-source harnesses reveal which control points matter most: memory, channels, terminal build loops, or automations.

Next 30 days

Evidence-backed value claims

The newsletter should elevate wins with named workflows, baselines, and verification methods, not generic productivity claims.

Changelog

Backfilled Agent Techniques Weekly issue 05 for Week 21 of 2026.