valid-review

name	valid-review
description	Review a valid model with inspect, lint/readiness, and coverage. Use when the user wants a capability summary, solver compatibility notes, and prioritized improvement advice.
argument-hint	[model-or-path] [--backend=BACKEND]
disable-model-invocation	true

Run a valid review workflow for $ARGUMENTS.

Target Resolution

1. Prefer direct MCP tools when a valid, pri, or similarly named verification server exposes inspect, lint or readiness, coverage, capabilities, or orchestrate.
2. If no matching MCP tool is available, fall back to CLI.
3. Resolve the target the same way as valid-check:
   • .valid file or existing path -> file mode
   • non-file first argument -> Rust registry model name
   • no argument with valid.toml -> resolve from cargo valid models --json or cargo run -q --bin cargo-valid -- valid models --json, then ask only when still ambiguous

CLI Fallback

• Rust/project mode:
  • Preferred if installed: cargo valid
  • Repo-local fallback: cargo run -q --bin cargo-valid -- valid
• .valid file mode:
  • Preferred if installed: valid
  • Repo-local fallback: cargo run -q --bin valid --
• Prefer --json for stable parsing.

Workflow

1. Run inspect first and capture the capability matrix.
2. Run lint/readiness.
   • Project mode: use readiness if available; it maps to lint-style readiness findings.
   • File mode: use lint.
3. Run coverage.
4. If the model has multiple properties and the user wants a broader health check, optionally run orchestrate to gather aggregate coverage.
5. Summarize solver compatibility from the capability matrix:
   • Whether the model is explicit-ready
   • Whether it appears IR/solver ready
   • Whether coverage, explain, and testgen are trustworthy for this model
6. Rank findings by severity and point at the smallest high-value fixes first.

Required Summary

Always include:

• Resolved target and execution path
• Capability matrix summary
• Solver/backend compatibility notes
• Readiness or lint findings
• Coverage strengths and gaps
• Prioritized recommendations

Prefer concrete examples such as missing guard coverage, explicit-only limitations, or properties that should be split or renamed.