---
variant: detail
kind: observations
slug: frontier-models-follow-complex-instructions
url: /observations/frontier-models-follow-complex-instructions
title: >-
  Frontier models such as GPT-5.5 and Claude Opus 4.7 are now good enough at
  detailed, complex instruction following that a well-structured harness can
  behave like a practical workflow engine.
source_path: content/observations/frontier-models-follow-complex-instructions.md
frontmatter:
  observation: >-
    Frontier models such as GPT-5.5 and Claude Opus 4.7 are now good enough at
    detailed, complex instruction following that a well-structured harness can
    behave like a practical workflow engine.
  observed: '2026-05-21T00:00:00.000Z'
  source: Tom
  context: >-
    Noticed while configuring an internal prototypes repository whose agent
    harness owns setup, GitHub checks, app configuration, commits, pushes, and
    validation.
  related_claims: []
  related_thoughts:
    - thoughts/agent-harnesses-are-state-machines
  related_projects:
    - projects/agent-harness
  tags:
    - agents
    - frontier-models
    - harnesses
    - workflows
links:
  related_thoughts:
    - status: resolved
      ref: thoughts/agent-harnesses-are-state-machines
      kind: thoughts
      slug: agent-harnesses-are-state-machines
      title: Agent harnesses are state machines
      url: /thoughts/agent-harnesses-are-state-machines
  related_projects:
    - status: resolved
      ref: projects/agent-harness
      kind: projects
      slug: agent-harness
      title: Agent Harness
      url: /projects/agent-harness
agent_metadata:
  source_path: content/observations/frontier-models-follow-complex-instructions.md
  html_url: /observations/frontier-models-follow-complex-instructions
  markdown_url: /observations/frontier-models-follow-complex-instructions.md
  source_url: >-
    https://github.com/flaming-codes/thinkinglabs/blob/main/content/observations/frontier-models-follow-complex-instructions.md
  summary: >-
    The model capability matters, but the harness is the multiplier. Detailed
    skills let broad instructions decompose into domain-specific operating
    procedures, so the agent can follow a long workflow without the top-level
    prompt becoming un...
  word_count: 222
  approx_token_count: 460
  token_estimate: chars/4
---
The model capability matters, but the harness is the multiplier. Detailed skills let broad instructions decompose into domain-specific operating procedures, so the agent can follow a long workflow without the top-level prompt becoming unreadable.