Canonical: https://botbento.com/usecases/screen-citizen-built-scripts-before-engineer-review/
Format: Markdown representation of the public HTML page.

# Screen citizen-built scripts before engineer review

[Coding](https://botbento.com/usecases/coding/) · Intermediate

Catch unsafe code and missing context before handoff

A bot watches a submission queue for scripts written by non-engineers. It checks each one for obvious hazards, missing tests, unclear ownership and risky data access, then prepares a short review note for an engineer. It keeps the queue organized and makes the handoff explicit instead of dropping unreviewed code into a backlog.

## Quick answers

**What does it do?** Catch unsafe code and missing context before handoff. Result: Review note for engineer.

**What does it need?** Tools: GitHub, Jira, Docs, Chat. Intermediate: some setup. The prompt includes an approval step.

**How do I run it?** When something happens: New script submitted. Copy the prompt below into a bot in BotBento.

## How it works

1. When something happens: New script submitted
2. Then: Static review and risk flagging
3. Result: Review note for engineer

## Prompt to start with

```text
You are a code intake reviewer. Monitor the submission queue for new scripts and inspect only the provided files. Summarize purpose, entry points, dependencies, data touched and test coverage. Flag hardcoded credentials, destructive commands, broad permissions, network calls and unclear ownership. Prepare a review note with severity, open questions and a recommended next action. Draft a reply to the submitter and a routing suggestion for an engineer, but wait for approval before sending, posting, merging, closing or deleting anything. Never execute untrusted code; analyze statically.
```

## Tools it uses

GitHub, Docs, Chat, Jira

## Works with

BotBento

## Tags

code-review, security, triage, workflow

## Source

[Goldman Sachs is pulling Claude Code from its bankers and traders](https://news.ycombinator.com/item?id=50415628) — Hacker News · checked 2026-11-13

Summarized in our own words. Any results described there are the authors’ own, not measured by BotBento.

Verified 2026-11-13: passed every automated gate and a second-model review (kimi-code/k3.1-pro). Automated, not human review. [How verification works](https://botbento.com/usecases/verification/)

By BotBento Editorial · Added 2026-11-13 · Updated 2026-11-13 · [Suggest a correction](https://botbento.com/contact/)

## Related use cases

- [Triage PHP bug reports into short fix plans](https://botbento.com/usecases/triage-php-bug-reports-into-short-fix/) — Reads each incoming issue, checks the code, and drafts a plan for review
- [Automated code security scanning and vulnerability checks](https://botbento.com/usecases/automated-codebase-security-scanning-and-exploit-verificatio/) — Scan source code for flaws and verify findings in a safe sandbox before filing tickets
- [Formal proof backlog triage for critical code modules](https://botbento.com/usecases/formal-proof-backlog-triage-for-critical-code/) — Keep unproven lemmas visible and route each proof gap to its owner
