# Human-held prompts

A person writes and edits the agent's instructions, skills and permissions; the agent doesn't write its own.

The agent's setup (its instructions, skills, rules, permissions and reference files) is kept in a [harness](/research/top-ai-agent-harnesses-survey) that a person edits by hand in a workbench. The agent can propose changes, but it can't apply them. Checks on the agent's [trace](/reference/glossary/trace) show whether it did what the harness asked, such as reading a given folder.

Our own runs: since 19 November 2026 our harnesses and loops are edited by hand in two workbench apps, and agents no longer touch them ([why letting Claude prompt itself fails](/posts/why-letting-claude-prompt-itself-fails)). Untested beyond our own runs; it is our own prediction that this guards against [memory ratcheting](/reference/failure-map/unattended-agent-failures/memory-ratcheting) and [security spazz outs](/reference/failure-map/unattended-agent-failures/security-spazz-out).

---

Published 2026-11-19. Web page: https://www.mackenziebowes.com/reference/failure-map/unattended-agent-failures/human-held-prompts
