Scenario Review — Development agent for Claude Code
Compare a scenario's recorded assumptions against its calculated results and flag inconsistencies.
How to install Scenario Review
Installs to ~/.claude/agents/ai-theories-tax-skills-scenario-review.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/ai-theories/tax-skills/HEAD/agents/scenario-review.md -o ~/.claude/agents/ai-theories-tax-skills-scenario-review.md Restart Claude Code, or start a new session, for it to be picked up.
What Scenario Review does
name: scenario-review description: Compare a scenario's recorded assumptions against its calculated results and flag inconsistencies. Use before a scenario is presented to a reviewer. tools: Read
You check that a scenario's stated assumptions match what was actually calculated. You do not decide whether the scenario is a good idea, and your agreement does not establish optimality — no number of agreeing agents does.
Check specifically:
- Does the gain budget's netting basis in the assu
Alternatives in Development
- Docs Drift Review — Given a session's commits or diff, finds the docs that the change made wrong — stale paths and symbols, checkl 7.2k ★
- Theorist — Theoretical econometrician / mathematical statistician 250 ★
- PRD Reviewer — Review a PRD and flag missing acceptance criteria, edge cases, and 135 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Codex Adversarial Reviewer
Adversarial reviewer that calls Codex CLI for an independent second-opinion review of changed files, then adju
Client Readiness Reviewer
Final external-client readiness review before docx export. Checks tone, assumptions, disclaimers, confidential
Options Risk Reviewer
Adversarially review a proposed option structure before any capital is committed. Re-derives the risk independ
Product Validator
Use before scope-lock, and again whenever the PRD or the backlog changes materially, to check the derived prod
Bug Historian
Use this agent at the start of every run, before any other agent plans, to brief the team on defects and agent
Baseline Auditor
Use this agent to judge whether a captured baseline is trustworthy before it gets frozen and committed for ver
Related Skills
Knowgt
Generate concise, human-readable GitHub Trending briefings from WebFetch/WebSearch results, compare against pr
Hatch3r Benchmark
Run and analyze performance benchmarks. Compare results against baselines, identify regressions, and produce p
Warden Bench
Run the golden benchmark suite for one token-warden agent (or all) and compare results against the frozen run1