Investigating A Codebase logo

Investigating A Codebase

Organization
ed3dai
investigating-a-codebase

Use when planning or designing features and need to understand current codebase state, find existing patterns, or verify assumptions about what exists; when design makes assumptions about file locations, structure, or existing code that need verification - prevents hallucination by grounding plans in reality

Overview

Publishered3dai
Repositoryed3d-plugins
Skill nameinvestigating-a-codebase
Stars
249
Forks
33
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by ed3dai on GitHub. Read the source before you install it.

Installation

Install the Investigating A Codebase AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/ed3dai/ed3d-plugins.git /tmp/ed3d-plugins
mkdir -p .claude/skills
cp -r /tmp/ed3d-plugins/plugins/ed3d-research-agents/skills/investigating-a-codebase .claude/skills/investigating-a-codebase
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Investigating A Codebase in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Investigating A Codebase on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Investigating A Codebase is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Investigating a Codebase

Overview

Understand current codebase state to ground planning and design decisions in reality, not assumptions. Find existing patterns, verify design assumptions, and provide definitive answers about what exists and where.

When to Use

Use for:

  • Verifying design assumptions before implementation ("Design assumes auth.ts exists - verify")
  • Finding existing patterns to follow ("How do we currently handle API errors?")
  • Locating features or code ("Where is user authentication implemented?")
  • Understanding component architecture ("How does the routing system work?")
  • Confirming existence definitively ("Does feature X exist or not?")
  • Preventing hallucination about file paths and structure

Don't use for:

  • Information available in external docs (use internet research)
  • Questions answered by reading 1-2 specific known files (use Read directly)
  • General programming questions not specific to this codebase

Core Investigation Workflow

Do not use nested subagents. If you are running as a research subagent, perform the investigation directly with Read/Glob/Grep and any other tools already available to you. Do not dispatch or invoke additional subagents.

  1. Start with entry points - main files, index, package.json, config
  2. Use multiple search strategies - Glob patterns, Grep keywords, Read files
  3. Follow traces - imports, references, component relationships
  4. Verify don't assume - confirm file locations and structure
  5. Report definitively - exact paths or "not found" with search strategy

Verifying Design Assumptions

When given design assumptions to verify:

  1. Extract assumptions - list what design expects to exist
  2. Search for each - file paths, functions, patterns, dependencies
  3. Compare reality vs expectation - matches, discrepancies, additions, missing
  4. Report explicitly:
    • ✓ Confirmed: "Design assumption correct: auth.ts:42 has login()"
    • ✗ Discrepancy: "Design assumes auth.ts, found auth/index.ts instead"
    • + Addition: "Found logout() not mentioned in design"
    • - Missing: "Design expects resetPassword(), not found"

Why this matters: Prevents implementation plans based on wrong assumptions about codebase structure.

Quick Reference

TaskStrategy
Where is XGlob likely names → Grep keywords → Read matches
How does X workFind entry point → Follow imports → Read implementation
What patterns existFind examples → Compare implementations → Extract conventions
Does X existMultiple searches → Definitive yes/no → Evidence
Verify assumptionsExtract claims → Search each → Compare reality vs expectation

Investigation Strategies

Multiple search approaches:

  • Glob for file patterns across codebase
  • Grep for keywords, function names, imports
  • Read key files to understand implementation
  • Follow imports and references for relationships
  • Check package.json, config files for dependencies

Don't stop at first result:

  • Explore multiple paths to verify findings
  • Cross-reference different areas of codebase
  • Confirm patterns are consistent not one-off
  • Follow both usage and definition traces

Verify everything:

  • Never assume file locations - always verify with Read/Glob
  • Never assume structure - explore and confirm
  • Document search strategy when reporting "not found"
  • Distinguish "doesn't exist" from "couldn't locate"

Reporting Findings

Lead with direct answer:

  • Answer the question first
  • Supporting details second
  • Evidence with exact file paths and line numbers

Provide actionable intelligence:

  • Exact file paths (src/auth/login.ts:42), not vague locations
  • Relevant code snippets showing current patterns
  • Dependencies and versions when relevant
  • Configuration files and current settings
  • Naming, structure, and testing conventions

Handle "not found" confidently:

  • "Feature X does not exist" is valid and useful
  • Explain what you searched and where you looked
  • Suggest related code as starting point
  • Report negative findings prevents hallucination

Common Mistakes

MistakeFix
Assuming file locationsAlways verify with Read/Glob before reporting
Stopping at first resultExplore multiple paths to verify findings
Vague locations ("in auth folder")Exact paths (src/auth/index.ts:42)
Not documenting search strategyExplain what was checked when reporting "not found"
Confusing "not found" typesDistinguish "doesn't exist" from "couldn't locate"
Skipping design assumption comparisonExplicitly report: confirmed/discrepancy/addition/missing
Reporting assumptions as factsOnly report what was verified in codebase

Frequently asked questions

What does the Investigating A Codebase AI skill do?

Use when planning or designing features and need to understand current codebase state, find existing patterns, or verify assumptions about what exists; when design makes assumptions about file locations, structure, or existing code that need verification - prevents hallucination by grounding plans in reality

Why use Investigating A Codebase on TypingMind?

Because you install it once and use it with any model. Investigating A Codebase is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Investigating A Codebase in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/ed3dai/ed3d-plugins/tree/main/plugins/ed3d-research-agents/skills/investigating-a-codebase. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Investigating A Codebase?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Investigating A Codebase?

As many as you like. As long as a model supports skills, you can use Investigating A Codebase with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Investigating A Codebase AI skill free?

It is published on GitHub by ed3dai. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇