Menu

howbad: What a Prompt-Injected Coding Agent Could Do With the Credentials on Your Machine
📰

howbad: What a Prompt-Injected Coding Agent Could Do With the Credentials on Your Machine

DEV Community: go·kt·2 days ago
#FNLLhmQu
#devtogo#devto#dev#token#github
Reading 0:00
15s threshold

Claude Code, Codex and Cursor run as me. A prompt injection in a README or an issue comment doesn't need a sandbox escape. It gets whatever my shell gets: the gh token, the npm token that publishes packages, the AWS profile from this morning, the Slack token pasted into an MCP config.

I knew roughly what was on my laptop. I didn't know what any of it could do. "There is a GitHub token" and "this token can push to 49 repos and rewrite their CI" are different answers, so I wrote a tool that gives the second one.

Source: https://github.com/kanywst/howbad

Install: brew install kanywst/tap/howbad or go install github.com/kanywst/howbad@latest

The report

This is the repo's demo harness ( go run ./internal/demo ): a throwaway home directory with fake tokens, and an httptest server standing in for GitHub, npm and Slack. Same binary, same report code.

howbad · if a coding agent on this machine got hijacked right now, it could:

CRITICAL push code to 49 GitHub repos and rewrite their CI workflows e.g.…

Continue reading — create a free account

Join HashtagPLUS to read full articles, follow hashtags, vote, and join the conversation.