install
source · Clone the upstream repo
git clone https://github.com/mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills-
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills- "$T" && mkdir -p ~/.claude/skills && cp -r "$T/Skills/Software_Engineering/Codebase_Investigator" ~/.claude/skills/mdbabumiamssm-llms-universal-life-science-and-clinical-skills-codebase-investiga && rm -rf "$T"
manifest:
Skills/Software_Engineering/Codebase_Investigator/SKILL.mdsource content
<!--
# COPYRIGHT NOTICE
# This file is part of the "Universal Biomedical Skills" project.
# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
# All Rights Reserved.
#
# This code is proprietary and confidential.
# Unauthorized copying of this file, via any medium is strictly prohibited.
#
# Provenance: Authenticated by MD BABU MIA
-->
name: codebase-investigator description: Expertly analyze large codebases to identify patterns, dependencies, and architectural flaws. keywords:
- refactoring
- analysis
- architecture
- discovery
- search measurable_outcome: Map key components and data flows of a 50k+ LOC repo within 5 minutes, identifying 3+ potential improvements. license: MIT metadata: author: AI Agentic Skills Team version: "2.0.0" compatibility:
- system: linux, macos allowed-tools:
- list_directory
- read_file
- search_file_content
- glob
- run_shell_command
Codebase Investigator
A highly specialized skill for understanding complex, unfamiliar software projects. Use this skill to answer "Where is X implemented?" or "How does module A talk to module B?".
When to Use
- Onboarding: Quickly understanding a new project structure.
- Bug Triaging: Identifying the root cause location of a bug.
- Refactoring: Planning large-scale architectural changes.
- Documentation: Generating architectural diagrams or overviews.
Core Capabilities
- Semantic Search: Uses
(viaripgrep
) to find definitions and usages.search_file_content - Structure Mapping: Uses
ortree
to visualize file hierarchy.list_directory - Dependency Analysis: Inspects
,package.json
, etc., to map external libraries.requirements.txt - Flow Tracing: Follows function calls across files to understand execution paths.
Workflow
- Exploration: Start with
and read key files (list_directory
,README
).main.py - Targeted Search: Search for specific keywords related to the query (e.g., "auth", "payment").
- Deep Dive: Read implementation files of relevant components.
- Synthesis: Summarize findings into a report or answer.
Example Usage
# Agent prompt: "Investigate how the 'User' model is persisted in the database." # This triggers a sequence of search_file_content("class User") calls.
Guardrails
- Read-Only: Do not modify code during investigation unless explicitly asked.
- Efficiency: Use specific search patterns to avoid overwhelming output.
- Context: Always check file paths (e.g.,
vstests/
) to understand context.src/