Agent Friendly Code

vercel-labs/deepsec

GitHub
https://github.com/vercel-labs/deepsec

Deepsec is a security harness for finding vulnerabilities in your codebase powered by coding agents

Stars:
8.3k
Language:
TypeScript
Default branch:
main
Last scored:
Overall

At a glance

Rank
#122 of 401
Top 30% · #35 of 82 TypeScript repos
Checks passed
9 of 17
Full list below
Best set up for
Pi
Score 79.9
Needs most work for
Aider
Biggest fix: .aider.conf.yml (+14.1 pts)

What it has

What it's missing

What to fix first

Pick an AI agent

Showing fixes for Claude Code

  1. 1
    Type configuration
    Add a type config (tsconfig.json for JS/TS, mypy.ini or pyrightconfig.json for Python). Rust/Go/JVM/Scala/Swift/C#/OCaml/Haskell/Zig are typed by default.
    +7.9 pts
  2. 2
    Pre-commit / git hooks
    Set up pre-commit (.pre-commit-config.yaml), husky, or lefthook to run format+lint on every commit.
    +5.3 pts
  3. 3
    Reproducible dev env
    Add a Makefile or devcontainer or Dockerfile so the agent can set up the project in one command.
    +4.7 pts

Score for each AI agent

Each agent weighs the checks by its own docs. Under each score are its top three. How each agent is scored →

Pi
#101 of 401
79.9
  • ✓AGENTS.md / CLAUDE.md
  • ~Test suite
  • ✓Linter / formatter config
Codex
#111 of 401
79.5
  • ✓AGENTS.md / CLAUDE.md
  • ✓README
  • ~Test suite
Kimi Code CLI
#107 of 401
79.3
  • ✓AGENTS.md / CLAUDE.md
  • ~Test suite
  • ✓Dependency manifest
Claude Code
#142 of 401
78.2
  • ✓AGENTS.md / CLAUDE.md
  • ~Test suite
  • ~Reproducible dev env
Devin
#107 of 401
76.5
  • ~Reproducible dev env
  • ✓Dependency manifest
  • ~Test suite
OpenHands
#108 of 401
72.3
  • ✓CI configuration
  • ✓Dependency manifest
  • ~Reproducible dev env
Gemini CLI
#138 of 401
69.0
  • ✗GEMINI.md
  • ✓README
  • ~Test suite
Cursor
#141 of 401
68.9
  • ✗Cursor rules (.cursor/rules)
  • ✓README
  • ✗Type configuration
Aider
#179 of 401
67.6
  • ✗.aider.conf.yml
  • ✓Linter / formatter config
  • ~Test suite

Does vercel-labs/deepsec have agent instructions?

An AGENTS.md (or CLAUDE.md) at the root tells AI coding agents how to build, test and work in the repo. A few agents also read a file of their own.

AGENTS.md / CLAUDE.md
CLAUDE.md (new tab) · 1,855 characters
Agent-specific files
.agents/skills (new tab) (1 skill)No GEMINI.md, .cursor/rules, .openhands/setup.sh or .aider.conf.yml
Repos that have one
62 of 82 TypeScript repos · 221 of 401 repos on the leaderboard · See examples →

All 17 checks · 9 passed

Skills (.agents/skills)· 100% pass
1 skill in .agents/skills/
.agents/skills/technical-writer/SKILL.md
AGENTS.md / CLAUDE.md· 100% pass
Substantive (1,855 characters)
CLAUDE.md
.aider.conf.yml· 0% pass
No .aider.conf.yml at repo root
CI configuration· 100% pass
5 GitHub Actions workflows
.github/workflows
CONTRIBUTING guide· 100% pass
Guide present
CONTRIBUTING.md
Cursor rules (.cursor/rules)· 0% pass
No .cursor/rules/*.mdc or .cursorrules
Dependency manifest· 100% pass
Manifest present
package.json
Reproducible dev env· 60% pass
package.json has 14 scripts
package.json
GEMINI.md· 0% pass
No GEMINI.md at repo root
License file· 100% pass
License present
LICENSE
Linter / formatter config· 100% pass
Config detected
biome.json
.openhands/setup.sh· 0% pass
No .openhands/setup.sh
Pre-commit / git hooks· 0% pass
No pre-commit / husky / lefthook found
README· 100% pass
README detailed (7,861 characters)
README.md
Manageable size· 100% pass
467 files
Test suite· 70% pass
Test files detected (e2e/bundle.test.ts)
e2e/bundle.test.ts
Type configuration· 0% pass
No type config found

How vercel-labs/deepsec compares

vs leaderboard average
7.8 pts above
Average 66.8
vs TypeScript average
3.4 pts above
Average 71.2
Stars among TypeScript repos
#71 of 82

Missing here, and how many repos have it

  • Type configuration · passed by 68% of leaderboard repos, 68% of TypeScript repos
  • Pre-commit / git hooks · passed by 29% of leaderboard repos, 48% of TypeScript repos
  • Cursor rules (.cursor/rules) · passed by 4% of leaderboard repos, 13% of TypeScript repos

Passes checks most repos miss

  • Skills (.agents/skills) · only 17% of repos pass it
  • Manageable size · only 16% of repos pass it

One fix for the most agents: Type configuration would raise the score for all 9 agents, by up to 11.5 points.

Similar repos

Same language and host as this repo, closest in overall score.

All TypeScript repos →

Add a badge to your README

Paste this at the top of the README. It shows the score, and links back to this page.

Agent friendly score for github/vercel-labs/deepsec
[![Agent Friendly](https://agentfriendlycode.com/api/badge/github/vercel-labs/deepsec.svg)](https://agentfriendlycode.com/repo/github/vercel-labs/deepsec)

Check the score on every pull request

A GitHub Action posts one comment per pull request with how the score changed and why.