---
title: "Test generator: review, role & definition · Undominated.ai"
canonical: https://undominated.ai/agents/trailofbits-6-test-generator/
description: "Pins the zeroize-audit test generator that stops when the final findings report is missing and still writes an empty manifest when nothing is confirmed."
---

# Test generator: review, role & definition · Undominated.ai

> Pins the zeroize-audit test generator that stops when the final findings report is missing and still writes an empty manifest when nothing is confirmed.

[← Explore all agents](/agents/)

SECURITY REVIEW / Trail of Bits

# Test generator

Pins the zeroize-audit test generator that stops when the final findings report is missing and still writes an empty manifest when nothing is confirmed.

 [Use this definition ↓](#setup)[Original source ↗](https://github.com/trailofbits/skills/tree/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/zeroize-audit/agents)

SOURCE REVIEW

 Reviewed 2026-10-14
 Evidence 4 linked sources
 Publisher Trail of Bits
 Licence [CC-BY-SA-4.0 ↗](https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/LICENSE)
 Revision 8e54050ae63c
 [Read what was—and wasn’t—checked ↓](#review)

“A harness writer that stops when the final findings report is missing and still writes an empty manifest when no finding is confirmed.”

 [Trail of Bits · upstream description ↗](https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/zeroize-audit/agents/6-test-generator.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - The agents index has no entry for plugins/zeroize-audit/agents/6-test-generator.md.
- The file treats a missing final report as fatal and treats zero confirmed findings as a non-error that still writes an empty manifest and Makefile.

### A good fit for

 - A finished zeroize-audit findings file, when the orchestrator asks for harness sources and a Makefile rather than another analysis pass

### Weigh up before choosing

 - This review did not execute the resource.
- A Markdown profile does not register a native subagent by itself.
- The install line adds the trailofbits/skills marketplace. It does not install the zeroize-audit plugin or this agent by itself.
- The file writes harness sources and a Makefile. This review did not generate or run them, and it does not repeat those steps.
- The plugin README marks this pass optional. The agent file itself does not say to skip the pass when the orchestrator does not call it.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - plugins/zeroize-audit/agents/6-test-generator.md
- plugins/zeroize-audit/README.md
- README.md
- LICENSE

### Our findings

 - The inputs are workdir, compile_db, config_path, final_report, and baseDir, all received from the orchestrator.
- The agent keeps findings whose confidence is confirmed or likely. It does not describe a path for any other confidence.
- A missing final report is fatal. The agent writes that error to notes and does not generate tests.
- Zero confirmed findings is not an error. The agent writes an empty manifest and a note.
- It always writes test_manifest.json and a Makefile, including a no-op pair. The process does not say to invoke that Makefile.

### Not established by this review

 - The resource was not executed and its tools were not called.
- No findings report was supplied, and no harness or Makefile was generated.
- The compile database and merged config named as inputs were not opened.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Use the role in your project.

[Upstream setup instructions ↗](https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/README.md)

DOCUMENTED COMMAND

 /plugin marketplace add trailofbits/skills Copy command ↗

Copying does not execute this command. It may retrieve a newer version than the reviewed source.

 - The README's Claude Code section is that line, then /plugin menu to browse plugins.
- The same README names zeroize-audit as the plugin for missing or compiler-eliminated secret zeroization in C, C++, and Rust. The marketplace command does not install that plugin or this agent by itself.
- The Codex section uses codex plugin marketplace add trailofbits/skills and a placeholder plugin name, so that line is not install.command.

### Before you start

 - Claude Code or Codex marketplace support
- The orchestrator inputs named in the file, including a final findings report

### Compatibility

Claude Code · Codex

### Implementation

c

### Harness writer that stops when the final findings report is missing

 - Read the final findings report, the merged config, and the compile database path
- Write harness files, a Makefile, and a manifest under the run directory
- Stop when the final report is missing, and do not treat zero confirmed findings as a failure

### Cost model

CC-BY-SA-4.0; no price is stated for this agent, and it was not run

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [6-test-generator ↗](https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/zeroize-audit/agents/6-test-generator.md) Checked 2026-10-14 https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/zeroize-audit/agents/6-test-generator.md Supports: fatal stop on a missing final report, and an empty manifest when nothing is confirmed
- [zeroize-audit README ↗](https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/zeroize-audit/README.md) Checked 2026-10-14 https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/zeroize-audit/README.md Supports: this agent is the optional runtime-harness pass
- [Repository README ↗](https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/README.md) Checked 2026-10-14 https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/README.md Supports: marketplace install line, Codex compatibility, and the zeroize-audit row
- [Repository LICENSE ↗](https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/LICENSE) Checked 2026-10-14 https://github.com/trailofbits/skills/blob/8e54050ae63c2f62224c35b1616efe71865b4777/LICENSE Supports: Attribution-ShareAlike 4.0 International

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Adversarial modeler ↗ Pins Trail of Bits' differential-review agent that models high-risk changes and says not to run on low or medium risk diffs.](/agents/trailofbits-adversarial-modeler/)[### C review worker ↗ Pins the c-review producing worker that reads source, writes one part file, and states that it has no shell.](/agents/trailofbits-c-review-worker/)[### Crypto Analyzer ↗ Reviews cryptographic design and code for weak primitives and mishandled keys, and points exploitation elsewhere.](/agents/pentest-crypto-analyzer/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Find reusable instructions](/skills/)
- [Inspect connections](/mcp-servers/)
- [Build a toolkit](/tools/)
- [Choose the model](/compare/)
