Enhancement: improve AI workflow with multi-agent pipeline and PR conventions by Miraeld · Pull Request #1068 · wp-media/imagify-plugin

Miraeld · 2026-05-25T00:50:45Z

Description

No linked issue — internal AI workflow improvement.

Brings the imagify-plugin .aiassistant workflow up to date with the same improvements applied to BackWPUp:

Multi-agent pipeline (grooming + lead review before push, QA after PR)
Hardened PR conventions (title format, Co-Authored-By, single Type of change, Made by AI label)
No-jQuery rule in wordpress-compliance
.claude/commands/ symlinks so Claude Code users get /skill-name slash commands

Type of change

Enhancement (non-breaking change which improves an existing functionality).

Detailed scenario

What was tested

Manually verified all skill and agent files for consistency. Symlinks confirmed to resolve correctly to .aiassistant/skills/*/SKILL.md (mode 120000 in git).

How to test

Invoke any skill via /issue-workflow, /imagify-architecture, etc. in Claude Code — resolves via symlink.
Run a new issue workflow — grooming-agent produces a spec before any code is written, lead-reviewer gates the push.
Verify git log on an AI-authored commit includes the Co-Authored-By trailer.

Affected Features & Quality Assurance Scope

AI workflow tooling only — no plugin source code changed.

Technical description

Documentation

New agents:

grooming-agent — analyses the issue and codebase before implementation, produces .TemporaryItems/.../issues/<N>-spec.md. Forces architectural reasoning (correct layer, root-cause vs workaround).
lead-reviewer — reviews the diff against the spec and project standards after commits, before push. Returns PASS or CHANGES REQUESTED with specific blockers.

Workflow changes (issue-workflow/SKILL.md):

Step 7: grooming-agent invocation replaces manual analysis
Steps 15–16: lead-reviewer gate with fix loop, before push
PR title formalised: Closes #N: title
Co-Authored-By trailer required on every AI-authored commit
Type of change: exactly one checkbox
Made by AI label applied on PR creation when available
Base branch is dynamic (defaults to origin/develop, overridable) and propagated to all agents

.claude/commands/ symlinks:
Each skill symlinked as a .md file — Claude Code users get /skill-name slash commands with no file duplication. Git tracks these as mode 120000.

wordpress-compliance:
Added no-jQuery rule — use native DOM APIs instead.

New dependencies

None.

Risks

None — no plugin source code touched.

Mandatory Checklist

Code validation

I validated all the Acceptance Criteria. If possible, provide screenshots or videos.
I triggered all changed lines of code at least once without new errors/warnings/notices.
I implemented built-in tests to cover the new/changed code.

Code style

I wrote a self-explanatory code about what it does.
I protected entry points against unexpected inputs.
I did not introduce unnecessary complexity.
Output messages (errors, notices, logs) are explicit enough for users to understand the issue and are actionnable.

Unticked items justification

No built-in tests — these are markdown workflow instruction files, not executable code.

Additional Checks

In the case of complex code, I wrote comments to explain it.
When possible, I prepared ways to observe the implemented system (logs, data, etc.)
I added error handling logic when using functions that could throw errors (HTTP/API request, filesystem, etc.)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

- Replace manual analysis step with grooming-agent invocation - Add lead-reviewer gate after implementation, before push - Formalise PR title format (Closes #N: title) - Require Co-Authored-By trailer on every AI commit - Restrict Type of change to exactly one checkbox - Apply Made by AI label on PR creation when available - Make base branch dynamic and propagate it to lead-reviewer and qa-engineer Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

codacy-production · 2026-05-25T00:51:24Z

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

🟢 Coverage ∅ diff coverage · +0.00% coverage variation

Metric Results

Coverage variation ✅ +0.00% coverage variation (-0.10%)

Diff coverage ✅ ∅ diff coverage (50.00%)

View coverage diff in Codacy

Coverage variation details

Coverable lines Covered lines Coverage

Common ancestor commit (efb23ff) 13940 48 0.34%

Head commit (f21b7ae) 13940 (+0) 48 (+0) 0.34% (+0.00%)

Coverage variation is the difference between the coverage for the head and common ancestor commits of the pull request branch: <coverage of head commit> - <coverage of common ancestor commit>

Diff coverage details

Coverable lines Covered lines Diff coverage

Pull request (#1068) 0 0 ∅ (not applicable)

Diff coverage is the percentage of lines that are covered by tests out of the coverable lines that the pull request added or modified: <covered lines added or modified>/<coverable lines added or modified> * 100%

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer}
_{TIP This summary will be updated as you push new changes.}

…agent Add architectural check (step 3c) prompting the grooming-agent to question whether a buggy method belongs in its current class before proposing a fix. Relabel former c check to d. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

lead-reviewer no longer embeds imagify-architecture and wordpress-compliance rules inline. Agent now reads the skill files at runtime, making skill files the single source of truth. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

- Add manager agent: reads spec, decides scope (patch vs refactor), dispatches to backend and/or frontend agents — asks user when scope is ambiguous - Add backend-agent: implements PHP changes, runs PHPCS + PHPStan - Add frontend-agent: implements JS/CSS changes, runs linting - Add imagify-frontend-architecture skill with frontend coding rules - Add .claude/commands symlink for new frontend architecture skill - Update grooming-agent: surfaces both implementation options without concluding; spec now includes an Implementation Options section for the manager to act on - Update issue-workflow: new 7-agent pipeline with max 3 retries on backend-agent, frontend-agent, lead-reviewer, and qa-engineer Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Replace flat manager pattern with an orchestrator that spawns specialized agents: grooming-reviewer, ci-agent, release-agent. Implementation agents (backend, frontend) now commit atomically. QA returns structured output. issue-workflow skill simplified to entry point only. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

- Replace grooming-reviewer with challenger (conditional, adversarial, APPROVED/NEEDS_REVISION/BLOCKED verdicts, MoSCoW findings, GitHub comment) - Orchestrator: risk-based CHALLENGER trigger, DOD L2 independent gate, criticality-based lead-reviewer routing, NTH dispatch, updated escalation rules - lead-reviewer: criticality tiers (CRITICAL/HIGH/MEDIUM/LOW), inline PR comments - backend-agent / frontend-agent: reframe Verify as DOD L1 with self-correction Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

…ent timeline Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Miraeld and others added 4 commits May 25, 2026 02:25

feat(claude): add .claude/commands symlinks and gitignore local settings

bdcf12c

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(compliance): enforce no-jQuery rule in wordpress-compliance skill

7247af5

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(agents): add grooming-agent and lead-reviewer sub-agents

c754cdf

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Miraeld self-assigned this May 25, 2026

Miraeld marked this pull request as ready for review May 25, 2026 00:51

Miraeld and others added 7 commits May 25, 2026 03:44

chore(agents): v3 orchestrator — adaptive routing, JSON contracts, ev…

35eddad

…ent timeline Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

fix(qa-agent): always boot env before testing, never skip on unreachable

f21b7ae

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Enhancement: improve AI workflow with multi-agent pipeline and PR conventions#1068

Enhancement: improve AI workflow with multi-agent pipeline and PR conventions#1068
Miraeld wants to merge 11 commits into
developfrom
enhancement/ai-workflow-improvements

Miraeld commented May 25, 2026

Uh oh!

codacy-production Bot commented May 25, 2026 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

Miraeld commented May 25, 2026

Description

Type of change

Detailed scenario

What was tested

How to test

Affected Features & Quality Assurance Scope

Technical description

Documentation

New dependencies

Risks

Mandatory Checklist

Code validation

Code style

Unticked items justification

Additional Checks

Uh oh!

codacy-production Bot commented May 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Up to standards ✅

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

codacy-production Bot commented May 25, 2026 •

edited

Loading