Daniel's personal website

This library entry is part of The Extended Frontier thesis. Entries are curated with AI assistance and human review; most initial entries were prepared with Claude (Anthropic), while individual entries may note other assisting systems. Metadata and annotations are editorial, not peer-reviewed. Entries flagged as unverified may contain placeholder dates, authors, or classifications.

Skill Issue: Harness Engineering for Coding Agents

HumanLayer··blog·source
Metadata unverified. URL, title, and publisher verified. Date is best-estimate from search snippet ('published in March 2026'); confirm the exact date and specific author(s) from the post header.
Skills, MCP servers, sub-agents, hooks, and back-pressure mechanisms are tactical solutions HumanLayer has arrived at.

Case-study framing of harness engineering for coding agents, with specific claims about what does and does not work (notably: role-based sub-agents don't work; sub-agents for context control do).

Classification

Role
case-study
Domain
software
Source type
blog
Harness types
execution-harnessrepair-harnessmonitoring-harnessinterface-harness
Validation position
during-generationimmediately-after-generationpost-deployment
Validation mode
empiricalmechanical
Prescription stance
strongly-procedural
Relation to argument
capability-is-extendedrepairability-mattersobservability-mattersbreakdown-when-harness-absent
Tags
harness-engineeringcoding-agentssub-agentscontext-controlback-pressure

Extended capability commentary

Input legibility
Task structure
Breaking work into discrete delegated tasks is a first-class move here.
Reward richness
Feedback latency
Repairability
Back-pressure mechanisms are repair harness by another name.
Observability
Offline evaluability
Institutional ratification

Why it matters

A strong counter-example to thin-harness-in-the-limit. HumanLayer has shipped coding-agent product and reports that sub-agents, hooks, and back-pressure do real work. Sharpens the disagreement with Tan/Miessler and localises it.

Annotation

HumanLayer's post is the library's best current counterweight to the thin-harness pole. The claim is not that more harness is always better — they explicitly report that role-based sub-agents ("frontend engineer," "backend engineer") don't work. The claim is that specific harness moves — sub-agents as context-control, hooks, back-pressure — carry real load and cannot be absorbed into a better model.

The piece is useful for the library because it:

  • Distinguishes harness types that work from those that don't, empirically rather than in principle.
  • Names specific mechanisms (sub-agents-for-context, back-pressure) that belong on the harness_types taxonomy.
  • Speaks from shipped product, which raises its weight on the practitioner-note vs framework-piece axis.

Read alongside

Related entries

  • What Is an Agent Harness
    Aparna Dhinakaran · 2026-04-21
    #coding-agentscapability-is-extendedrepairability-mattersobservability-mattersbreakdown-when-harness-absentexecution-harnessrepair-harnessmonitoring-harnessinterface-harness
  • Good and Bad Harness Engineering
    Daniel Miessler · 2025-08-31
    #harness-engineeringcapability-is-extendedrepairability-mattersobservability-mattersbreakdown-when-harness-absentexecution-harnessrepair-harnessmonitoring-harness
  • LLM Knowledge Bases
    Andrej Karpathy · 2026-04-01
    capability-is-extendedrepairability-mattersobservability-mattersexecution-harnessrepair-harnessmonitoring-harnessinterface-harness
  • An open-source spec for Codex orchestration: Symphony
    Alex Kotliarskyi, Victor Zhu, and Zach Brock · 2026-04-26
    capability-is-extendedrepairability-mattersobservability-mattersexecution-harnessrepair-harnessmonitoring-harnessinterface-harness

Overlap is computed on tags, relation-to-argument, and harness types — not on role or domain, because contrasts are often the most useful neighbours.