Skip to content

FLAB: streamline test permissions and PPE write defaults - #109

Merged
MonaDevAI merged 1 commit into
mainfrom
fix/large-source-edit-review
Nov 6, 2026
Merged

MonaDevAI merged 1 commit into
mainfrom
fix/large-source-edit-review

Conversation

@MonaDevAI

Copy link
Copy Markdown
Contributor

Problem

Routine tests, PPE writes and oversized source edits can stall on repetitive or unavailable approvals.

Change

Default dashboard/MCP routine access and configured PPE writes, with opt-outs. Add active-session checkpoint reads and guided, explicitly authorized fresh-edit recovery without replaying denials.

Validation

  • Focused tests pass
  • npm pack --dry-run contains the required files
  • Documentation is updated
  • No credentials, tokens, private endpoints, or test data are included

Windows: 497 unit tests passed. Full browser run: 49/51; fixed the missing-identity resume regression. Both failed cases and both new flows passed serial reruns (4/4). Final lint, types and formatting passed. The original full run remains failed.

Risk

PPE defaults authorize configured origins only; grants expire. Authentication and managed policy remain enforced. No npm release.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 46e778f9-d964-46b0-bebe-3b6f9f71eb74
Copilot AI balanced review requested due to automatic review settings November 6, 2026 15:44
@MonaDevAI
MonaDevAI merged commit 89fee92 into main Nov 6, 2026
5 checks passed

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

🟡 Changes recommended

Edited or stale prepared authorizations can be sent without review IDs, bypassing server binding checks and allowing jobs to resume with changed or stale authorization.

1 open issue
What changed in this PR

This PR reduces repetitive approval prompts in FixLab’s dashboard and connected MCP workflows, and adds guided recovery for oversized source edits.

Changes:

  • Default routine job access and configured PPE writes, with explicit opt-outs.
  • Allow bounded reads of active-session checkpoints and add separately authorized fresh-edit recovery.
  • Add regression coverage and document permission defaults and recovery boundaries.
File Description
test/​dashboard-server.test.js Tests connected defaults and bound edit recovery.
test/​dashboard-runtime-requests.test.js Tests recovery metadata and denial preservation.
test/​dashboard-runtime-policy.test.js Tests oversized-edit eligibility and checkpoint restrictions.
mcp/​dashboard-client.js Applies routine-access and configured PPE defaults.
e2e/​dashboard-readiness.spec.js Covers PPE opt-outs and guided recovery interactions.
docs/​security.md Documents PPE defaults and consent boundaries.
docs/​commands.md Explains permission defaults and recovery flows.
docs/​architecture.md Describes the updated permission and recovery model.
dashboard/​server.js Exposes readiness and manages bound recovery state.
dashboard/​runtime-requests.js Validates recovery metadata and generates authorization text.
dashboard/​runtime-permission-policy.js Adds checkpoint reads and oversized-edit eligibility checks.
dashboard/​public/​index.html Adds recovery controls and updates write guidance.
dashboard/​public/​app.js Implements PPE defaults and guided edit-recovery interactions.
dashboard/​dev-post.js Adds PPE configuration inspection.
dashboard/​copilot-session.js Passes session identity and eligible recovery metadata.

Review effort: Balanced

Comment thread dashboard/public/app.js
Comment on lines +2971 to +2974
...(sourceEditAuthorizationSelection && sourceEditAuthorizationSelection.instanceId === dashboardInstanceId &&
sourceEditAuthorizationSelection.jobId === activeJob?.id &&
details === sourceEditAuthorizationSelection.authorization
? {
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