2026-07-068 min read
The Terminal Wakes Up
A fictional agent opens its day in a project directory, then the margin notes explain what Claude Code actually can and cannot do.
Read entrySerialized field notes
Diary entries
Each entry has two layers: the fictional diary voice, and the source-backed margin that explains which public documents support the factual claims.
Freshness surface
This index is designed for weekly additions. New entries should extend the diary format while keeping source notes current and linking sibling topics outward instead of duplicating them.
2026-07-069 min read
When the Context Window Fills
The agent feels the session getting heavy; the source notes explain context windows, compaction, and why bigger context is not automatically better.
Read entry2026-07-0610 min read
The Moment Before Permission
A fictional agent pauses before acting; the annotations explain tool calls, approval, auto-approval, and oversight patterns.
Read entry2026-07-068 min read
A Memory Made of Files
The agent finds a memory in files and protocols; the notes explain Claude Code instructions and MCP without turning this into a memory tutorial.
Read entry2026-07-069 min read
Refusing the Wrong Shortcut
A first-person scene about saying no to a harmful or brittle path, anchored in Anthropic alignment and character sources.
Read entry2026-07-068 min read
After the Transcript
A session ends; the annotations explain privacy and retention sources without turning the page into legal advice.
Read entry2026-07-0610 min read
The Model Learns a Character
A reflective entry about assistant voice, grounded in Anthropic character and persona-selection research.
Read entry2026-07-0611 min read
Why Fiction Belongs in Safety Work
The entry argues for careful AI fiction, then anchors that claim in Anthropic research on fictional stories and alignment.
Read entry