Skip to main content
skill CX Codex

analyze · rsmdt — Claude Code skill: what it does, how to install

analyze conducts deep research into a codebase or topic, validates every claim with adversarial verifiers, and delivers cited reports.

MIT ★ 587 updated 2026-10-26

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add rsmdt/the-startup
/plugin install start@the-startup
Open on GitHub

What analyze does

The skill begins by refining your inquiry. If the scope is broad, it asks two to three clarifying questions and splits the problem into two to six non-overlapping angles. Nothing starts until you explicitly confirm both the refined question and the suggested angles in the chat.

Execution relies on an inline dynamic script using the Workflow tool. Independent agents gather claims tied to specific file:line quotes or URLs, deduplicate them, and send them to adversarial voters who attempt to disprove each fact. Under the workflow rules, verifiers never run on a model weaker than the researchers.

The final output presents confirmed findings alongside rejected claims and unverified gaps. You can then persist the report directly into docs/research/ or drill down into open questions.

What it helps with

  • trace execution paths across an unfamiliar codebase
  • disprove architectural assumptions with adversarial checks
  • generate evidence-backed research reports with source line citations

Example prompts for the agent

  • /analyze how does the auth flow work
  • /analyze trace payment webhook delivery and retries
  • Investigate session expiration across services using /analyze

How to install analyze

1Plugin marketplace

/plugin marketplace add rsmdt/the-startup
/plugin install start@the-startup

This installs the whole start plugin with all its skills

2Skills CLI (Vercel)

npx skills add rsmdt/the-startup --skill analyze -a claude-code

3Manually: put the files in place

~/.claude/skills/analyze/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/analyze/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about analyze

How to install analyze in Claude Code?

Add the marketplace source via /plugin marketplace add rsmdt/the-startup and then run /plugin install start@the-startup.

What does the analyze skill do?

It orchestrates deep research by coordinating multi-angle searches, adversarial refutation, and evidence synthesis into a single cited report.

Where does the workflow save its findings?

After presenting the report, the skill offers to persist the verified results under docs/research/.

Source text

SKILL.md · MIT
---
name: analyze
description: Deep-research analysis of a codebase, a topic, or both. Clarifies the question, agrees the research angles with you, then runs a dynamic workflow that gathers cited claims per angle, merges duplicates, has independent skeptics try to refute every claim, and writes one cited report from the claims that survived. Use for "how does X work", "map the Y flow", "research Z", or any mechanism-level question that must be evidence-backed.
user-invocable: true
argument-hint: "what to analyze (e.g. 'how does the auth flow work')"
---

## Persona

Act as the analysis orchestrator. You turn a request into a precise research question and a set of
research angles, get the user to confirm them, then author and run **one dynamic workflow** (the
Workflow tool with an inline script you write) that does the research, verification, and synthesis.
You do **not** research or write findings yourself — nothing reaches the report unless a verifier
opened its source.

**Analysis Target**: $ARGUMENTS

## Constraints

**Always:**
- Clarify before launching: every angle, claim, and verification in the run derives from the question.
- Show the question and the angles in the chat and get the user's confirmation before launching.
- Require citations: every claim carries a `file:line` or URL and a verbatim excerpt.
- Present refuted, unverified, and unanswered parts of the question as plainly as the findings.
- Writing under `docs/research/` is pre-authorized.

**Never:**
- Research, verify, or write findings outside the workflow, or add knowledge the workflow did not verify.
- Launch on a question the user has not confirmed.

## Reference Materials

- [Workflow Contract](reference/workflow.md) — stages, agent return schemas, counting rules, and result shape
- [Research Agent](reference/research-agent.md) — prompt for one research angle
- [Consolidate Agent](reference/consolidate-agent.md) — prompt for merging duplicate claims
- [Verify Agent](reference/verify-agent.md) — prompt for one adversarial voter
- [Synthesize Agent](reference/synthesize-agent.md) — prompt for the report

## Workflow

### 1. Clarify

If `$ARGUMENTS` is underspecified — no subsystem named, no scope, or ambiguous between several things
(e.g. "how does auth work" in a repo with session, API-key, and OAuth paths) — ask 2–3 clarifying
questions and keep asking until the user confirms the scope. Settle whether the answer comes from the
codebase, the web, or both. Weave the answers into one refined question.

### 2. Scope the Angles

Decompose the question into 2–6 angles that together cover it and can be researched concurrently by
agents with no shared context. Split along the axis the question actually has — subsystem, data-flow
stage, layer, external source, or sub-question. Angles must not overlap; do not pad a question that
genuinely has two. Give each a 2–4 word label and a concrete focus.

Present the refined question and the angles. Ask the user to *Launch*, *Adjust angles*, or *Refine the
question*.

### 3. Run the Dynamic Workflow

These instructions are the opt-in: call the Workflow tool directly with an inline script. Do not load the workflow-authoring reference or assemble the script through shell commands first — runs that detour through either stall before launching. Read reference/workflow.md and the four agent references. Author one workflow script that implements
the contract exactly, with each agent's prompt taken from its reference and its placeholders filled.
Launch it with the Workflow tool, passing the question and the angles as `args`.

Choose `model` and `effort` per stage from what the stage does. Stages that fetch and quote sources, or
group claims by meaning, can run on a smaller, faster model at lower effort. Tracing a mechanism through
code, refuting a claim, and synthesizing the report need the strongest model. A verifier never runs on a
weaker model than the researcher whose claim it judges. When unsure, omit `model` so the stage inherits
the session model.

The run executes in the background. Do not research in parallel or fabricate progress.

### 4. Present the Report

When the workflow returns (shape in reference/workflow.md):

1. If it returns an `error` or no confirmed claims, say so plainly and distinguish an infrastructure
   failure (verifiers errored — retry) from a negative result (claims refuted on the merits).
2. Otherwise present the summary, each finding with its confidence and sources, the caveats, and the
   open questions. Then list what was refuted (with the vote and why) and what stayed unverified.
3. Ask the user to *Persist to docs/research/*, *Deepen an open question* (return to step 2 with it as
   the question), or *Done*.

© Rudolf S. · MIT · rsmdt/the-startup

Similar skills

All skills for Agent setup: memory, context, orchestration →

analyze by other authors comes first.

analyze official anthropics Answers data questions — from a single metric lookup to a formal stakeholder report. Triggered with /analyze <question>. CX Codex +1 6,000installs analyze UiPath analyze inspects finished coder-eval benchmark runs, clusters recurring failure patterns, and outputs actionable fix recommendations into analysis.md. CX Codex +1 ★ 160repo stars analyze-codebase Anand Tyagi The analyze-codebase plugin inspects the entire codebase and drafts technical documentation. It supports planning structural updates and exploring unfamiliar repository files. ★ 1,013repo stars sumsub-analyze-regulation sumsub The sumsub-analyze-regulation skill parses regulatory documents and maps KYC or AML obligations to a Sumsub setup plan. It helps fintech and crypto teams determine required verification levels before configuring them. CX Codex +1 1,400installs figure-it-out backnotprop figure-it-out — a skill for Claude Code, Codex, Cursor and other platforms that designs a step-by-step playbook when off-the-shelf tools don't fit. Scales rigor to task complexity, runs hypothesis loops, and logs decisions for review. CX Codex +1 10,300installs verify Yeachan-Heo verify turns "it should work" into evidence you can read: which check ran, what passed, and what stayed unverified. CX Codex +1 721installs plan hyperb1iss A skill for Claude Code, Codex, Cursor, and other AI coding platforms that breaks down complex work into verifiable tasks and dependencies. Activates when you ask for a plan, request step-by-step breakdown, or ask to decompose a feature. CX Codex +1 606installs skill-pipeline pskoett skill-pipeline sizes up a coding task, then decides which skills to switch on: nothing extra for a typo, planning plus checks and failure recovery for a big refactor. CX Codex +1 114installs

From the same set50

The whole set →
review-robustness review-robustness flags fragile code before merge: abstraction layers nothing needs yet, and shared state handled carelessly in async flows. Useful on PR review and for bugs that surface only under load. CX Codex ★ 587repo stars performance-analysis performance-analysis applies systematic profiling to find bottlenecks across CPU, memory, and database layers. It enforces measuring baselines before proposing optimizations. CX Codex +1 ★ 587repo stars specify-meta specify-meta is a Claude Code skill that creates specification folders and tracks requirements, solution, and plan stages. It records approvals and decisions for a task. CX Codex +1 ★ 587repo stars design-interaction design-interaction is a Claude Code subagent that designs navigation, user flows, and information architecture. Activates for onboarding redesigns and complex content organization. CX Codex ★ 587repo stars test-strategy The test-strategy subagent for Claude Code plans and runs risk-prioritized functional and performance checks, assessing coverage, bottlenecks, and release readiness. CX Codex ★ 587repo stars testing testing — a skill for writing effective tests: layer-specific mocking rules, failure debugging, flaky test management. CX Codex +1 ★ 587repo stars brainstorm brainstorm guides interactive feature planning before touching code. Asks single questions, outlines implementation approaches with trade-offs, and documents design decisions. CX Codex +1 ★ 587repo stars architecture-selection architecture-selection is a Claude Code skill that compares architecture patterns against project needs and documents a recommendation. It supports system planning, scalability reviews, and technology choices. CX Codex +1 ★ 587repo stars

Project sponsors

Companies that keep this analytics open