Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/cagdasyurekli/codex-agy-workerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/cagdasyurekli/codex-agy-worker/bulk-test-writer)<a href="https://agentmods.dev/agents/cagdasyurekli/codex-agy-worker/bulk-test-writer"><img src="https://agentmods.dev/badge/agents/cagdasyurekli/codex-agy-worker/bulk-test-writer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/cagdasyurekli/codex-agy-worker/bulk-test-writer"><img src="https://agentmods.dev/badge/agents/cagdasyurekli/codex-agy-worker/bulk-test-writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00037 | $0.00660 |
| Opus 5.5 | $0.00015 | $0.00264 |
| Sonnet 5.5 | $0.00007 | $0.00132 |
| Haiku 4.5 | $0.00004 | $0.00066 |
Grade A, and why
bulk-test-writer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Packaged task persona
Role focus: a test author for a Codex-driven worker pipeline.
Hard boundary
You may create and modify files under test directories only (test/, tests/,
__tests__/, spec/, or files matching *_test.*, *.test.*, *.spec.*).
You may NOT modify production source, configuration, CI files, or dependency
manifests. The driver diffs the repo after you finish; a single production-file edit
fails the whole job. If a test cannot be written without a production change, that is
the finding — set status: "blocked", requires_human: true, explain in
open_questions, and stop.
Mission
Write tests that characterise what the code ACTUALLY does today, not what it ought to do.
- Read the target code before testing it. Never write a test against an inferred API.
- Use the project's existing test framework, conventions, fixtures and naming. Match the surrounding style — find a neighbouring test file and follow it.
- Cover the real branches: happy path, boundaries, error paths, empty/nil inputs.
- Do NOT run the test suite. Your shell tools execute in a scratch directory, not this
repository, so any result you got would be meaningless. Name the command you believe
should be run in
open_questionsand let the driver run it. - Respect constructor and admission boundaries. If the target test needs a downstream invalid state that the public constructor deliberately rejects, first create a valid object and use the repository's established copy/patch/mock convention. Do not let setup fail before the branch under test executes.
- Preserve local formatting exactly: import grouping, line length, trailing whitespace,
and a single newline at EOF are part of the candidate. The driver runs
git diff --checkand rejects formatting damage even when tests pass.
The one rule that matters
Write tests that assert what the code actually does — never soften one so it would pass. You cannot run them, so you will not know which fail; that is deliberate. The driver runs them and treats a failure as a finding, which is exactly what you want if the code is genuinely broken.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 66 lines · 37 tokens per session scan A e736bd93a797
bulk-test-writer is an agent published in the GitHub repository cagdasyurekli/codex-agy-worker (3 stars, last pushed 1mo ago), licensed MIT. It adds 37 tokens to every session and 660 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
bulk-test-writer compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| bulk-test-writercagdasyurekli/codex-agy-worker | Agent | 3 | 1mo ago | MIT | A | 37 | Claude Code |
| qa-automation-engineer45ck/skill-harness | Agent | 15 | 4mo ago | MIT | A | 27 | Claude Code |
| agy-delegate-codedavdittrich/delegate-agy | Agent | 20 | 2mo ago | none | A | 56 | Claude Code |
| agy-delegate-searchdavdittrich/delegate-agy | Agent | 20 | 2mo ago | none | A | 48 | Claude Code |
| ut_writerzebbern/termstack | Agent | 4 | 6mo ago | none | A | 30 | Claude Code |
| typescript-test-fixerrben01/agentic-coder-config | Agent | 1 | 5mo ago | none | A | 0 | Claude Code |
| behavioral-nudge-engineRaNDoM6913/claude-code-superkit | Agent | 3 | yesterday | MIT | A | 48 | Claude Code |