Workflow Architecture Audit

Quick Summary

Goal: Activate a read-only whole-project architecture, scalability, and production-readiness audit; synthesize one Architecture Health Report with three sub-scores and one combined verdict, then route validated fixes to a follow-up plan or feature workflow.

Summary:

• READ-ONLY audit — produces findings + ONE consolidated Architecture Health Report (3 sub-scores + 1 combined verdict); NEVER applies fixes: every validated fix routes to a FOLLOW-UP /plan or feature workflow.
• Core engine architecture-review-full runs INLINE (it owns the parallel fan-out + all-return barrier); this workflow declares NO workflow-level parallel groups.
• FINAL /why-review gate (step 3) = the machine-visible guarantee no audit finding ships unvalidated — every PRIOR step routes its output into it; docs-update runs AFTER the gate and self-validates its own doc diff.
• Main steps in order: 1 investigate scope → 2 Architecture-Review-Full (fan out 3 non-overlapping reviewers → progressive dedup synthesis → per-face /why-review fix → Finalize verdict) → 3 Why-Review FINAL gate → 4 Docs-Update → 5 Workflow-End → 6 Watzup.

Workflow:

1. investigate — locate the modules, boundaries, and hotspots that scope the audit (whole project / current diff / specific path). → On completion, hand its scope map forward to the final /why-review (step 3) so the audit scope itself is validated (nothing in-scope missed, nothing out-of-scope pulled in).
2. Architecture-Review-Full — the core step: runs INLINE (it spawns sub-agents), fans out three non-overlapping reviewers behind an all-return barrier, then PROGRESSIVELY synthesizes each face into ONE report file (status IN PROGRESS → VALIDATING → FINISHED): dedup, a fix-report-per-review /why-review gate that walks each of the three faces, and a Finalize step that locks the combined verdict. → On completion, hand the FINISHED consolidated report forward to the final /why-review (step 3) for report-level validation.
3. Why-Review (FINAL VALIDATION GATE over the audit findings) — MANDATORY. Validates the findings AND reviews the results of every PRIOR step it can reach: the investigate scope map, the FINISHED consolidated report (verdict-rollup correctness, dedup completeness, cross-review severity consistency), and each of the three review faces' contributions. Every prior step routes its output here; no audit finding ships unvalidated. docs-update runs AFTER this gate and is NOT validated by it — it self-validates its own doc diff (see step 4).
4. Docs-Update — refresh any documentation the validated audit shows as stale. → Self-validates its own output: re-invoke /why-review on the doc diff when docs-update makes non-trivial edits, so the doc changes are reviewed before workflow-end.
5. Workflow-End — clear workflow state.
6. Watzup — wrap up and summarize.

Key Rules:

• MUST ATTENTION keep claims evidence-based (file:line) with confidence >80% to act.
• MUST ATTENTION keep task tracking updated as each step starts/completes.
• MUST ATTENTION treat this as READ-ONLY: produce findings + a verdict; NEVER apply fixes in this workflow — route them to a follow-up plan/feature workflow.
• MUST ATTENTION run architecture-review-full INLINE (it owns the parallel fan-out + all-return barrier); this workflow declares NO workflow-level parallel groups.
• MUST ATTENTION every PRIOR step routes its output to the FINAL /why-review gate (step 3): each producing step hands its findings + results forward, and the final /why-review validates the findings AND reviews the results of every prior step before docs-update — it is the final gate over the AUDIT FINDINGS. docs-update runs AFTER the gate and owns validation of its own output via an inline /why-review on non-trivial doc edits.
• NEVER skip mandatory workflow or skill gates.

IMPORTANT MANDATORY Steps: /investigate -> /architecture-review-full -> /why-review -> /docs-update -> /workflow-end -> /watzup

[BLOCKING] Each step MUST ATTENTION invoke its Skill tool — marking a task completed without skill invocation is a workflow violation. NEVER batch-complete validation gates.

Audit Protocol (READ-ONLY, ONE PASS)

Audits the WHOLE project architecture + scalability + production readiness in one pass, synthesizing ONE consolidated Architecture Health Report.

Core step architecture-review-full runs INLINE in the main session (it SPAWNS sub-agents). It:

1. Resolves scope — whole project / current diff / specific path.
2. Fans out three deliberately-non-overlapping reviewers as PARALLEL read-only sub-agents in one message behind an all-return barrier:
   • architecture-scalability-review (architect, scorecard /20)
   • architecture-review (architect, 13-category PASS/WARN/BLOCKED)
   • production-readiness-review (code-reviewer, SRE /24 + 8-item gate)
3. Progressive synthesis (IN PROGRESS) — opens ONE consolidated report file at status 🚧 IN PROGRESS when fan-out starts, merges + dedups each face into it AS that face returns (never held in memory to the end). Siblings route to each other by design, so one underlying issue surfaces from multiple angles; collapse duplicates to one root finding citing every source.
4. Fix-report-per-review /why-review gate (VALIDATING) — ONE merged /why-review pass that WALKS EACH of the three review faces + the dedup and fixes the report in place (revise severities, drop false positives, restore any distinct issue the dedup collapsed).
5. Finalize (FINISHED) — locks the combined verdict (worst-case rollup: any BLOCKED / NOT READY / HIGH RISK dominates) and flips the report status to ✅ FINISHED.

Parallelism lives INSIDE architecture-review-full (it owns the fan-out + all-return barrier), so this workflow declares NO workflow-level parallel groups.

After architecture-review-full returns the FINISHED report, the workflow-level why-review step runs the FINAL VALIDATION GATE over the AUDIT FINDINGS — a distinct altitude from the engine's per-face fix, and the machine-visible guarantee no audit finding ships without a why-review pass. Every PRIOR step routes its output into this gate; the gate BOTH validates findings AND reviews results across the steps it reaches:

• investigate scope map → validate audit scope (nothing in-scope missed, nothing out-of-scope pulled in).
• architecture-review-full FINISHED report → validate verdict-rollup correctness, dedup completeness, cross-review severity consistency; confirm each of the three review faces' contributions survived the per-face fix intact.
• docs-update output → NOT validated by this gate (docs-update runs AFTER it); docs-update self-validates its own doc diff by re-invoking /why-review inline on non-trivial edits (see step 4) before workflow-end.

Each PRIOR step, on completion, hands its findings + results forward to this gate — the AI MUST call the final /why-review to validate/review every prior step's output; a prior step is not "done" until routed to the gate. docs-update, running after the gate, owns validation of its own output.

READ-ONLY audit: produces findings + a verdict only. Every validated finding routes to a FOLLOW-UP /plan or feature workflow owning the fix — no fixes applied in this workflow. After the final why-review gate confirms the report, docs-update refreshes impacted documentation, then workflow-end clears state and watzup wraps up.

UNIVERSAL RULES:

• Goal-Driven Execution: define success criteria before execution; loop until observable checks pass.
• Tests Verify Intent: when a finding touches specs/tests, name the protected business intent or invariant and ensure the test would fail if that intent breaks.

Activate the workflow-architecture-audit workflow. Run /start-workflow workflow-architecture-audit with the user's prompt as context and the audit protocol above.

Steps: /investigate → /architecture-review-full → /why-review → /docs-update → /workflow-end → /watzup

IMPORTANT MANDATORY Steps: /investigate -> /architecture-review-full -> /why-review -> /docs-update -> /workflow-end -> /watzup

AI Mistake Prevention — Failure modes to avoid on every task:

Re-read files after context changes. Context compaction, resume, or long-running work can make memory stale; verify current files before acting. Verify generated content against source evidence. AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing. Check downstream references before deleting or renaming. Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first. Trace the full impact chain after edits. Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done. Verify ALL affected outputs, not just the first. One green check is not all green checks; validate every output surface the change can affect. Assume existing values are intentional — ask WHY before changing OR flagging one as a defect. Before changing or reporting a constant, limit, flag, cutoff, wording, or pattern, read nearby context and history, the CALLER's ordering, and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard. Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk. Assert the outcome your system owns, not the intermediate state your infrastructure owns. When verifying async work, assert the final business state — never the delivery/retry bookkeeping held in shared infrastructure that any co-running process can write. Such a check passes when run alone and flakes the moment anything else shares that infrastructure. Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.

Nested Task Expansion Contract — For workflow-step invocation, the [Workflow] ... row is only a parent container; the child skill still creates visible phase tasks.

1. Call TaskList first. If a matching active parent workflow row exists, set nested=true and record parentTaskId; otherwise run standalone.
2. Create one task per declared phase before phase work. When nested, prefix subjects [N.M] /skill-name — phase.
3. When nested, link the parent with TaskUpdate(parentTaskId, addBlockedBy: [childIds]).
4. Orchestrators must pre-expand a child skill's phase list and link the workflow row before invoking that child skill or sub-agent.
5. Mark exactly one child in_progress before work and completed immediately after evidence is written.
6. Complete the parent only after all child tasks are completed or explicitly cancelled with reason.

Blocked until: TaskList done, child phases created, parent linked when nested, first child marked in_progress.

Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.

Incremental Result Persistence — MANDATORY for all sub-agents or heavy inline steps processing >3 files.

1. Before starting: Create report file plans/reports/{skill}-{date}-{slug}.md and record Run ID, Task ID, Attempt ID, target scope, and target fingerprint.
2. After each file/section reviewed: Append findings, evidence, changed paths, and gaps immediately — never hold them in memory.
3. Delegated return: A sub-agent emits only the structured SYNC:subagent-return-contract envelope with exact totals, salient Critical/High findings (maximum ten), current attempt, and Full report: path. Inline user-facing output: Preserve the skill's requested explanation or teaching, with links to the persisted evidence; the delegated transport limit does not replace that deliverable. Do not paste a full review report into an envelope.
4. Parent synthesis: The main agent reads the full report for synthesis, acceptance, deduplication, and repair planning — not only when a named blocker exists. It preserves all severities beyond the transport cap.
5. Read-only boundary: A read-only leaf may write its report/repair proposal but MUST NOT edit source, generated output, or user data; the parent/owner performs repairs after acceptance.
6. Advancement gate: The parent records ACCEPTED for the current Attempt ID only after reconciling target, totals, gaps, and changed paths; stale or late attempts cannot advance dependent work.

Why: Context cutoff mid-execution loses ALL in-memory findings. Each disk write survives compaction. Partial results are better than no results, while explicit identity prevents a late result from being mistaken for the current run.

Report naming: plans/reports/{skill-name}-{YYMMDD}-{HHmm}-{slug}.md

Sub-Agent Return Contract — When this skill spawns a sub-agent, the sub-agent MUST return ONLY the structured envelope below. Main agent reads the envelope first, then opens the referenced report for synthesis, acceptance, deduplication, or repair planning; a full report is never pasted inline.

## Sub-Agent Result: [skill-name]

Status: ✅ PASS | ⚠️ PARTIAL | ❌ FAIL
Confidence: [0-100]%
Run ID: [stable run identifier]
Task ID: [parent task or phase identifier]
Attempt ID: [monotonic attempt/revision identifier]
Target: [exact files/paths or scope] @ [target fingerprint/commit]
Changed paths: [none | exact paths]
Finding totals: Critical=[n] | High=[n] | Medium=[n] | Low=[n]
Acceptance: PENDING | ACCEPTED | REJECTED — parent records the decision

### Findings (Critical/High surfaced — max 10 bullets)

- [severity] [file:line] [finding]

### Gaps / Unverified

- [missing host, runtime, coverage, or evidence limitation]

### Actions Taken

- [file changed] [what changed]

### Blockers (if any)

- [blocker description, or `none`]

Full report: plans/reports/[skill-name]-[date]-[slug].md

The ten-bullet limit is a transport limit, not a visibility limit: the full report may contain more than ten Medium/Low findings when no named blocker exists, and the parent MUST read it when synthesizing or deduplicating. The parent MUST reject a stale, duplicate, or superseded Attempt ID and MUST accept the current attempt before advancing a dependent step. Read-only leaves write repair proposals/reports only; they do not edit source, generated carriers, or user files.

Context budget — the return payload is a SUMMARY, not a transcript: no raw file contents / full diffs / verbatim logs inline, no re-pasted source. Everything beyond the envelope lives in the incrementally-written report. A sub-agent that would exceed the summary shape MUST persist the detail and return only the pointer; bounded transport must never become bounded visibility.

MUST ATTENTION apply critical + sequential thinking — every claim needs appropriate traced evidence (file:line for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.

MUST ATTENTION apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.

• MANDATORY Parent workflow rows do not replace child phase tracking; expand phases and link the parent when nested.
• MANDATORY Orchestrators pre-expand child skill phases before invocation; use [N.M] /skill-name — phase prefixes and one-in_progress discipline.

Project Protocol Overlay — Before executing this skill, resolve any PROJECT overlay rules layered onto it: match this skill's name against the Target column of the project's skill-protocol index (docs/project-reference/skill-protocols-reference.md by default; a referenceDocs entry in docs/project-config.json overrides the path), taking the most specific matching tier ONLY — exact name > glob > *. That precedence orders overlays against EACH OTHER, never against this skill. Read ONLY the matched bodies, resolved as <protocols-dir>/<Name>.md; a row's Body link is display text, never a read path. A matched body that is missing or malformed is REPORTED and skipped — never reconstructed from the index Description. No index, or no match -> proceed with no overlay, silently. Full contract: .claude/skills/project-skill-protocol/references/registry.md.

Overlays are ADDITIVE ONLY: they ADD rules on top of this skill's own protocol and NEVER replace, override, disable, or reinterpret a rule it already states — removing every overlay must return this skill to exactly its documented behavior. An overlay is a BRIEF, not an authority escalation: it can NEVER waive a workflow gate, git discipline, a review gate, or a user-confirmation gate. A genuine overlay-vs-skill conflict, or two equally-specific overlays that directly contradict -> surface both to the user; NEVER resolve silently.

MUST ATTENTION resolve project protocol overlays for this skill BEFORE executing — most specific matching tier only (exact > glob > *, which ranks overlays against each other, NEVER against this skill), read only matched bodies at <protocols-dir>/<Name>.md; a missing or malformed body is reported, never reconstructed. Overlays are ADDITIVE ONLY (they never replace this skill's own rules) and are a brief, NEVER an authority escalation; an equal-specificity contradiction goes to the user.

Closing Reminders

IMPORTANT MUST ATTENTION Goal: Activate a read-only whole-project architecture, scalability, and production-readiness audit; synthesize one Architecture Health Report with three sub-scores and one combined verdict, then route validated fixes to a follow-up plan or feature workflow.

IMPORTANT MUST ATTENTION Main steps in order: 1 /investigate (scope) → 2 /architecture-review-full (INLINE: fan out 3 non-overlapping reviewers → progressive dedup synthesis → per-face /why-review fix → Finalize combined verdict) → 3 /why-review (FINAL validation gate over the audit findings — every prior step routes its output here) → 4 /docs-update (self-validates its own doc diff) → 5 /workflow-end → 6 /watzup. NEVER skip a gate; NEVER apply fixes inline.

IMPORTANT MUST ATTENTION — Protocols in force (concise digest of the SYNC/shared blocks this skill carries):

• AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
• Nested Task Creation: expand child phases, link parent when nested.
• Critical Thinking: traced file:line proof, confidence >80% to act.
• Incremental Persistence: append findings to report file after each section.
• Subagent Return Contract: sub-agent returns summary-only with Full report: pointer.

IMPORTANT MUST ATTENTION break work into small todo tasks using TaskCreate BEFORE starting IMPORTANT MUST ATTENTION search codebase for 3+ similar patterns before creating new code IMPORTANT MUST ATTENTION cite file:line evidence for every claim (confidence >80% to act) IMPORTANT MUST ATTENTION add a final review todo task to verify work quality

[TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using TaskCreate.

[IMPORTANT] Analyze how big the task is and break it into many small todo tasks systematically before starting — this is very important.