Skip to content

docs(red-team): follow-up fixes to Quick Start #325

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Aryansharma28 merged 1 commit into from
Apr 13, 2026
+3 −13
Merged

docs(red-team): follow-up fixes to Quick Start #325

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 3 additions & 13 deletions docs/docs/pages/advanced/red-teaming/quick-start.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,7 @@ Works with Claude Code, Cursor, Claude Desktop, Codex, and any MCP-compatible cl
}
```

Restart the client. Full reference: [LangWatch MCP Server](https://langwatch.ai/docs/integration/mcp).
Grab your API key from [app.langwatch.ai](https://app.langwatch.ai) → **Settings → API Keys**, then restart your client. Full reference: [LangWatch MCP Server](https://langwatch.ai/docs/integration/mcp).

## 3. Ask your assistant to generate the test

Expand Down Expand Up @@ -150,21 +150,11 @@ describe("Agent security", () => {

## 4. Run it

:::code-group

```bash [python]
pytest tests/red_team/ -v
```

```bash [typescript]
npm test -- tests/red-team
```

:::
Run your usual test command, or just ask your coding assistant to run it and monitor it for you. Red team runs are long — 50 turns can take several minutes and will consume real LLM tokens on both the attacker and target models — so make sure your runner's per-test timeout is generous.

Each turn prints the attacker's message, your agent's response, and a per-turn score. A failing test includes the full transcript and the judge's reasoning — you see exactly which turn broke the agent and how.

## 5. View the run in LangWatch (optional)
## 5. View the run in LangWatch

If you've instrumented your agent with LangWatch, every red team run appears in the Simulations dashboard: full attack transcripts, per-turn scores, and side-by-side comparison across runs to track whether a prompt change made your agent more or less resilient.

Expand Down
Loading