Advanced Strategic-Planning Criteria

Compatibility note: this file keeps the historical docs/v7-scope.md path and V7-linked internal identifiers for scripts, tools, and tests. Public product language should describe this as an advanced strategic-planning maturity layer, not as a product version.

This criteria layer remains local-first. It excludes hosted service work, database/storage work, GitHub/PR adapters, CLI UX, and typed SDKs.

It should make architect-mcp better at strategic planning, local simulation, standards negotiation, and self-improvement while staying inside the existing MCP/domain/test architecture.

Strategic-Planning Direction

This layer focuses on local strategic intelligence:

• Architecture strategy maps.
• Standards negotiation.
• Local what-if planning.
• Agent behavior diagnostics.
• Rule authoring assistance.
• Quality trend snapshots without persistence.
• Human-readable governance packs.

Architecture Strategy Maps

Goal: help agents understand how project goals, risks, stack choices, standards, and verification connect.

Deliverables:

• Strategy-map output from a project brief and selected packs.
• Risk-to-standard mapping.
• Workflow-to-boundary mapping.
• Verification-to-risk mapping.
• Gaps where important risks have no standard, pack, or verification check.

Standards Negotiation

Goal: make tradeoffs explicit when strict standards would slow or over-constrain the project.

Deliverables:

• Compare minimal, balanced, and strict standards profiles.
• Explain what each profile catches and what it allows.
• Suggest one focused question when standards conflict with user speed preferences.
• Mark non-negotiable safety rules separately from adjustable quality rules.

Local What-If Planning

Goal: preview likely review outcomes before an agent writes code.

Deliverables:

• What-if review for proposed file plans plus selected standards profiles.
• Impact preview for adding/removing a pack or policy bundle.
• Blast-radius forecast from proposed files, change type, and standards.
• Verification forecast: checks likely required before claiming completion.

Agent Behavior Diagnostics

Goal: identify recurring agent failure patterns from supplied session summaries and review outputs without storing history.

Deliverables:

• Diagnose patterns such as over-broad edits, skipped verification, premature root-cause claims, tool overuse, repeated assumptions, and ignored contracts.
• Return corrective steering text for the next agent turn.
• Produce compact "watch for this" handoff notes.
• Score whether the agent followed the intended harness loop.

Rule Authoring Assistance

Goal: help maintainers write better local rules without live fetching or external systems.

Deliverables:

• Draft candidate rules from supplied source text, findings, or examples.
• Explain whether a rule is executable, manual-only, too vague, duplicated, or too broad.
• Suggest detector metadata and fixture pairs.
• Suggest rule names, finding codes, examples, and remediation text.

Quality Trend Snapshots

Goal: compare two explicitly supplied review reports without persistence.

Deliverables:

• Trend comparison between before/after reports.
• New, resolved, worsened, improved, and unchanged finding groups.
• Standards coverage delta.
• Verification honesty delta.
• Suggested next local eval fixture based on recurring findings.

Human-Readable Governance Packs

Goal: produce standards packs that humans can review before agents enforce them.

Deliverables:

• Markdown rendering for policy packs and stack packs.
• Plain-English summaries of rules, triggers, examples, and evidence.
• "Adopt now / trial / reject" recommendation blocks.
• Review checklist for maintainers approving a pack.

Non-Goals

• Hosted service or hosted API implementation.
• Database/storage, including schema, persistence, migrations, review history, or memory storage.
• GitHub repository, branch, PR, issue, app, OAuth, or comment workflow.
• CLI UX.
• Typed client SDK.
• Billing, accounts, teams, dashboards, or remote policy management.

Those remain out of scope until a later backend/productization phase.