Skip to main content
skill CX Codex

deep-dive · akillness — Claude Code skill: trace causes, then define requirements

deep-dive — a two-stage skill: three parallel lanes trace why something broke, then a requirements interview starts from those findings. Built for bugs and changes in code you don't know yet.

no license 177 installs ★ 61 updated 2026-11-19

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add akillness/jeo-skills --skill deep-dive -a claude-code
Open on GitHub

What deep-dive does

Deep Dive runs two stages back to back. The first digs into why something happens, splitting the work into three investigation lanes: a cause in the code, one in config or environment, and a mismatch between what is measured and what is assumed. The second stage converts what turned up into a spec through an interview.

Findings cross over at three injection points. The interview opens from a richer starting point, receives system context, and begins with the questions the trace could not settle on its own. State sits in the directory of whichever adapter is active (.omc/, .omx/, .agents/), and artifact paths are stored, so a run picks up again after the context gets compacted.

The skill does not execute anything itself; it hands the spec to the runtime's planner or executor. The handoff counts as ready once the saved files pass scripts/validate_deep_dive_artifacts.py. You confirm the lane hypotheses once at the start, and the trace then runs without interruption. Already know the root cause? Run /deep-interview on its own.

What it helps with

  • trace a bug whose cause you can't pin down
  • map how an existing subsystem behaves before changing it
  • carry investigation findings into a spec instead of restarting the interview
  • split a vague problem into three checkable hypotheses
  • hand a finished spec to the runtime's planner

Example prompts for the agent

  • Deep dive on the flaky password reset emails: find the cause first, then write up what we should change.
  • Run deep-dive on the queue worker crash, then turn the findings into requirements.
  • I want to know why the report export got slow before we plan any changes.

How to install deep-dive

1Skills CLI (Vercel)

npx skills add akillness/jeo-skills --skill deep-dive -a claude-code

2Manually: put the files in place

~/.claude/skills/deep-dive/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/deep-dive/

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

Questions about deep-dive

How do I install deep-dive in Claude Code?

Run npx skills add akillness/jeo-skills --skill deep-dive -a claude-code. The skill lands at ~/.claude/skills/deep-dive/SKILL.md.

What does deep-dive do?

It chains a cause-tracing stage with a requirements interview, so the second stage starts from what the first one found rather than from scratch.

Does deep-dive need a particular runtime?

One adapter per run: OMC under Claude, OMX under Codex, OMA under Gemini/Antigravity. The trace and spec schema stays the same, while orchestration maps to whatever primitives the current tool actually has.

Similar skills

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

deep-dive by other authors comes first.

deep-dive rohitg00 deep-dive runs multi-step deep research in Claude Code using DAG query planning, parallel subagents, and gap iteration without external APIs. CX Codex +1 ★ 2,858repo stars deep-dive ZaxbyHub deep-dive audits a selected part of a codebase without editing source files. Use it to investigate several aspects of a project and review evidence-backed findings. CX Codex +1 ★ 539repo stars nansen-wallet-deep-dive nansen-ai The nansen-wallet-deep-dive skill runs a wallet through labels, balance, PnL, transactions, perp positions and counterparties via the Nansen CLI. Useful when you need to know who is behind an address. CX Codex +1 393installs user-deep-dive PostHog user-deep-dive turns an email address into a PostHog user profile: what the person does in the product and which reports they keep open. Handy before a customer call. CX Codex +1 212installs anthropic-technical-deep-dive curiositech anthropic-technical-deep-dive builds deep technical preparation for Anthropic interviews: Constitutional AI, RLHF, interpretability, scaling laws, MCP, agentic systems, and AI safety. CX Codex +1 141installs keyword-deep-dive inhouseseo The keyword-deep-dive skill studies one keyword: reads the Google SERP, inspects the top three competitors and returns a 90-day ranking plan. Useful when you are deciding whether a query is worth chasing. CX Codex +1 ★ 416repo stars Animation Principles - Deep Dive dylantarre Animation Principles - Deep Dive — a reference skill covering Disney's 12 animation principles with technical detail: deformation ratios, frame counts, arcs, follow through. Useful when the agent needs animation context. CX Codex +1 560installs Redfin Property MCP — Single-Address Deep Dive nexgendata-apify Redfin Property MCP — Single-Address Deep Dive is an MCP server that searches Redfin listings by filters and retrieves a home's detailed history by address or URL. CX Codex +1 5,877installs

From the same set99

The whole set →
hyperfine-benchmarking hyperfine-benchmarking — a skill that runs commands with warmup cycles and collects timing statistics. Compare tool variants or validate performance changes before committing, with exportable evidence. CX Codex +1 251installs ghgrab The ghgrab skill lets an agent find and download chosen GitHub files, folders, and release assets from the terminal without cloning the whole repository. CX Codex +1 251installs semble The semble Claude Code skill searches code by behavior or symbol name. It activates when an agent needs to locate an implementation or find code similar to a known snippet. CX Codex +1 234installs react-bits Steers an agent through react-bits: pick dependencies, drop an animated component into a React project, fix GSAP and Three.js leaks, or send your own component upstream. CX Codex +1 211installs spec-kit Installs GitHub's specify CLI, bootstraps a spec-driven project for your agent, and runs the /speckit.* pipeline from constitution to implementation. CX Codex +1 167installs academic-research academic-research sorts a research topic or paper draft into one of several linked academic workflows, from literature discovery to peer review. The skill itself decides which branch and mode fit the request. CX Codex +1 148installs opik Configure Opik to trace LLM calls, evaluate responses, and monitor runs. Use the skill when you need to assess prompts or compare experiments. CX Codex +1 146installs heretic The heretic skill runs directional ablation and refusal analysis on open-weight models. Use it to study refusals or reduce benign over-refusal without fine-tuning. CX Codex +1 143installs

Project sponsors

Companies that keep this analytics open