Skip to content

task-observer: docs/ai/writing-docs updates from 2026-05-01#4

Open
divad12 wants to merge 1 commit intomasterfrom
task-observer/writing-docs-2026-05-01
Open

task-observer: docs/ai/writing-docs updates from 2026-05-01#4
divad12 wants to merge 1 commit intomasterfrom
task-observer/writing-docs-2026-05-01

Conversation

@divad12
Copy link
Copy Markdown
Owner

@divad12 divad12 commented May 1, 2026

Target\nglobal-doc: docs/ai/writing-docs.md\n\n## Observations addressed\n- Observation 2 (dotfiles/log.md): Rationale is part of the instruction for judgment-heavy rules → changed: added bullet to Contract section — when editing/compressing instructions, preserve short rationale for judgment-heavy rules (where model must choose how to frame output, not execute a mechanical command)\n- Observation 3 (dotfiles/log.md): Hard rules need failure-mode rationale alongside the imperative → changed: same bullet — preserve failure-mode context alongside hard rules (what breaks, what thought pattern is a stop signal, what action is allowed instead); stripping either reduces enforcement strength without reducing length


Generated by Claude Code

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant