Skip to content

Require necessity checks in agent plans and reviews - #953

Merged
taras merged 1 commit into
mainfrom
agent/require-justified-mechanisms
Oct 19, 2026
Merged

taras merged 1 commit into
mainfrom
agent/require-justified-mechanisms

Conversation

@taras

@taras taras commented Oct 19, 2026

Copy link
Copy Markdown
Owner

Why

Planning and review can approve unnecessary machinery when they check only whether implementation follows an accepted plan. Tests created around that machinery can reinforce the assumption without establishing a production need.

What changes

Before: the role guides require reuse and concrete problems, but do not explicitly separate a mechanism's necessity from plan compliance during review.

After: the Planner justifies new registries, protocols, lifecycle guards and generic interfaces before freezing a plan. Planner and Architect reviews independently assess necessity and return unsupported assumptions for correction. The Planner also applies the existing comment rule to the final diff.

How to verify it

  • git diff --check HEAD^ HEAD passes.
  • Manually reviewed both guides against their existing role boundaries and frozen-evidence rules.
  • Documentation only; no runtime tests needed.

Scope confirmation

  • Every changed file supports the purpose described above.
  • Unrelated cleanup and formatting changes are excluded.
  • No generated or mechanical changes are included.
  • The description matches the final diff and verification.

@github-actions

Copy link
Copy Markdown

PR #953: Require necessity checks in agent plans and reviews

2 files, +34 / -1

Scope

✅ PR scope looks good.

Structural

✅ No structural bloat detected.

Slop

✅ Slop indicators look low.

Static Analysis

✅ Oxlint found no issues.

Correctness

No extraneous code patterns detected.

@taras
taras marked this pull request as ready for review October 19, 2026 08:20
@taras
taras enabled auto-merge October 19, 2026 08:20
@taras
taras merged commit 15f4ffa into main Oct 19, 2026
164 of 170 checks passed
@taras
taras deleted the agent/require-justified-mechanisms branch October 19, 2026 12:26
@taras

taras commented Oct 19, 2026

Copy link
Copy Markdown
Owner Author

The security model and agreed createHooks contract are in follow-up #954 at a236889. #953 was already merged, so these additions are a separate documentation change against main.

The follow-up description records the implementation and cleanup work: implement the adapter with real consumers; assess Component.expand and bound-command observation first; classify APIs with intentional transformations or provider routing before conversion; and reconcile existing defensive specifications/tests with the approved threat model while preserving consequential reliability guarantees.

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