Skip to content

docs(architecture): align tests with compounding capability - #1100

Merged
drewstone merged 1 commit into
mainfrom
docs/compounding-capability-framing
Sep 5, 2026
Merged

docs(architecture): align tests with compounding capability#1100
drewstone merged 1 commit into
mainfrom
docs/compounding-capability-framing

Conversation

@drewstone

Copy link
Copy Markdown
Contributor

Repository guidance treated a single-run steering comparison as a prerequisite for learning across projects and testing richer agent designs. Align the guidance with the documented goal: capability that improves across projects, including better learning methods.

Centralize experiment scope in the architecture document. Permit tests of interacting mechanisms, require actual resource accounting, and retain explicit conditions for rejecting a mechanism. State what tests and scanners can establish without treating their results as complete proof.

Validation: pnpm run docs:check, final pnpm run docs:freshness, added-link checks, git diff --check, and independent review of all six changed files passed. No source or public API changes.

@drewstone
drewstone merged commit a16d8a3 into main Sep 5, 2026
4 checks passed
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