simplicio-review

Runs parallel adversarial subagent reviews of branch diffs and synthesizes a deduplicated merge verdict.

9|Updated May 8, 2026
One-click install
npx skills add https://github.com/simpletibr/simplicio-loop-marketing --skill simplicio-review-simpletibr
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: simplicio-review
Source: https://github.com/simpletibr/simplicio-loop-marketing/tree/main/.claude/skills/simplicio-review
Command: npx skills add https://github.com/simpletibr/simplicio-loop-marketing --skill simplicio-review-simpletibr

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Single-reviewer code reviews tend to rubber-stamp changes, letting security holes, unmet acceptance criteria, and fake placeholder returns slip into merged branches. ## Core Features & Use Cases - Parallel adversarial fan-out: Spawns three independent subagents in one message, each with a distinct rubric covering security/correctness, code quality, and runtime reproduction. - Evidence-gated verdicts: Requires web_verify screenshots for front-end changes, flow_audit evidence for cross-surface changes, and impact_audit coverage for shared-contract changes before passing. - Deduped synthesis: Merges findings by file:line and normalized claim, counts reviewer votes, and emits a machine-readable verdict of pass, fix-required, or block. - Use Case: Before merging a pull request that touches both a React component and its API endpoint, run this review to confirm the acceptance criteria actually reproduce end-to-end and that no injection or broken caller exists in the diff. ## Quick Start Ask the assistant to run a thermo-nuclear adversarial review on the current branch diff against the base branch before merging.

Frequently Asked Questions about simplicio-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run an adversarial code review on a branch before merging?

Invoke the review on the branch diff against its base, and it spawns three parallel subagents covering security/correctness, code quality, and runtime reproduction. Findings are deduplicated by file:line and synthesized into a pass, fix-required, or block verdict.

What does a multi-agent code review check that a single reviewer misses?

Independent reviewers on distinct rubrics catch logic bugs, injection and secret leaks, unmet acceptance criteria, fake placeholder returns, and structural quality issues. Overlapping findings across reviewers raise confidence through vote counting.

Does the review require extra evidence for front-end changes?

Yes. Diffs touching front-end files require a web_verify ledger entry with a screenshot, trace path, and zero console errors. Cross-surface changes additionally require flow_audit evidence, and shared-contract changes require impact_audit coverage.

Can this review scope itself to only the changed lines?

Yes. Scope is strictly added or modified lines in the diff; pre-existing issues are out of scope unless the change makes them reachable. This keeps findings confirmed, in-scope, and actionable.

When should a pull request be blocked instead of approved?

A majority-refute verdict on any acceptance criterion sends the change back for fixes, and a single high-confidence security finding blocks regardless of votes. Missing required runtime evidence also results in fix-required.