Skip to content

FE-698: Agent mutation audit; tool surface design#104

Merged
lunelson merged 4 commits into
mainfrom
ln/fe-698-slice-d
May 8, 2026
Merged

FE-698: Agent mutation audit; tool surface design#104
lunelson merged 4 commits into
mainfrom
ln/fe-698-slice-d

Conversation

@lunelson
Copy link
Copy Markdown
Contributor

@lunelson lunelson commented May 7, 2026

FE-698 audit agent mutation surface

FE-698 settle intent graph and changeset vocabulary

FE-698 document mutation operation vocabulary

@lunelson lunelson changed the title FE-698 audit agent mutation surface FE-698: Agent mutation audit; tool surface design May 7, 2026
@lunelson lunelson self-assigned this May 7, 2026
@lunelson lunelson requested a review from kostandinang May 7, 2026 16:23
@lunelson lunelson marked this pull request as ready for review May 7, 2026 16:23
@cursor
Copy link
Copy Markdown

cursor Bot commented May 7, 2026

PR Summary

Low Risk
Documentation-only changes that don’t affect runtime behavior; risk is limited to potential terminology drift or future implementation assumptions.

Overview
Adds a new design audit (docs/design/AGENT_MUTATION_SURFACE.md) that inventories current agent-originated/adjacent durable write paths, grades boundary strength, and proposes a canonical capability/mutation-surface operation vocabulary (authority classes, <noun>.<verb> ids, and a changeset-centered model for future intent-graph edits).

Updates memory/PLAN.md and memory/SPEC.md to align roadmap/spec language with the new terminology: intent graph/intent items and changeset/change (vs. knowledge/patch), and references the audit as input to the upcoming capability registry work.

Reviewed by Cursor Bugbot for commit 22d555c. Bugbot is set up for automated code reviews on this repo. Configure here.

@augmentcode
Copy link
Copy Markdown

augmentcode Bot commented May 7, 2026

🤖 Augment PR Summary

Summary: This PR documents Brunch’s current agent-originated mutation boundaries and aligns future-facing vocabulary around intent-graph and changeset-ledger semantics.

Changes:

  • Adds docs/design/AGENT_MUTATION_SURFACE.md, an audit of agent-originated/adja­cent durable write paths with authority class + boundary-quality classification.
  • Defines a semantic operation vocabulary (<noun>.<verb>) and maps current DB-helper primitives to proposed Brunch-owned handlers.
  • Calls out pressure points (observer graph writes, interviewer tools writing durable state, thin graph-edit routes) and recommends next implementation slices.
  • Updates memory/PLAN.md to treat “changeset/change” as the canonical mutation-ledger vocabulary (with “patch” as historical terminology).
  • Updates memory/SPEC.md to prefer “intent graph / intent items / intent edges” as the product vocabulary, while keeping knowledge_* as transitional implementation names.

Technical Notes: Documentation-only change; intended as design input for a future capability registry and server-owned mutation handler surface.

🤖 Was this summary useful? React with 👍 or 👎

Copy link
Copy Markdown

@augmentcode augmentcode Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Review completed. 2 suggestions posted.

Fix All in Augment

Comment augment review to trigger a new review at any time.

Comment thread memory/SPEC.md Outdated
Comment thread memory/SPEC.md Outdated
@lunelson lunelson force-pushed the ln/fe-698-slice-c branch from 396d945 to d138c56 Compare May 7, 2026 16:41
@lunelson lunelson force-pushed the ln/fe-698-slice-d branch from 138e633 to 3fef067 Compare May 7, 2026 16:41
Copy link
Copy Markdown
Contributor Author

lunelson commented May 7, 2026

@kostandinang docs/design/SIDE_CHAT.md and docs/superpowers/specs/2026-05-05-side-chat-highlight-to-note-design.md are still present at this point, should they be deleted?

Copy link
Copy Markdown
Contributor

kostandinang commented May 7, 2026

Thanks for flagging @lunelson, opened a PR with the cleanup for superpowers plan and archived the design one, as it may have some design notes that may need later.

kostandinang
kostandinang previously approved these changes May 8, 2026
@lunelson lunelson force-pushed the ln/fe-698-slice-d branch from 3fef067 to e431537 Compare May 8, 2026 11:02
@lunelson lunelson force-pushed the ln/fe-698-slice-c branch from d138c56 to 3f5a1fc Compare May 8, 2026 11:02
@graphite-app graphite-app Bot changed the base branch from ln/fe-698-slice-c to graphite-base/104 May 8, 2026 11:02
@lunelson lunelson changed the base branch from graphite-base/104 to ln/fe-698-slice-c May 8, 2026 11:02
@github-actions github-actions Bot dismissed kostandinang’s stale review May 8, 2026 11:02

Latest approval commit 3fef067 is not an ancestor of e431537, indicating rewritten history after approval

Base automatically changed from ln/fe-698-slice-c to main May 8, 2026 11:04
@lunelson lunelson force-pushed the ln/fe-698-slice-d branch from e431537 to 22d555c Compare May 8, 2026 11:04
@graphite-app
Copy link
Copy Markdown

graphite-app Bot commented May 8, 2026

Merge activity

  • May 8, 11:05 AM UTC: Graphite rebased this pull request, because this pull request is set to merge when ready.

@kostandinang kostandinang self-requested a review May 8, 2026 11:06
@lunelson lunelson added this pull request to the merge queue May 8, 2026
Merged via the queue into main with commit 9400c1e May 8, 2026
6 checks passed
@lunelson lunelson deleted the ln/fe-698-slice-d branch May 8, 2026 11:07
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.

2 participants