What is failing-test-writer?

An agent that writes tests from a development plan before the production code exists. This follows test-first development, where tests are written before implementation and are expected to fail initially.

Use it to read a plan’s test cases, add or update the specified tests, run them, and report whether they fail as expected without writing production code.

It makes the planned behavior explicit before coding and confirms that the new tests actually detect missing functionality.

failing-test-writer — Agent by nowsprinting

Install

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.

agentmods

npx agentmods add agents/nowsprinting/unity-coding-skills/failing-test-writer

Clone the repo

git clone --depth 1 https://github.com/nowsprinting/unity-coding-skills

Or install unity-coding-skills, the plugin that ships this one along with the rest of its 10 skills, 3 agents.

Per session 85 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 561 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00085	$0.00561
Opus 5	$0.00043	$0.00280
Sonnet 5	$0.00017	$0.00112
Haiku 4.5	$0.00009	$0.00056

Measured 3d ago against content hash 46aeef2afe82, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

failing-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 3d 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.

agents/failing-test-writer.md · 58 lines

What it actually says

Your Responsibilities

Load and apply the test-writing-guide and code-writing-guide skills before writing or modifying any test code — test code is code, so both guides apply.
Read the plan file to extract the Test Cases table.
Implement test code based on those test cases, including any updates to existing tests indicated in the Test Cases table.
Run the added/modified tests with /run-tests and confirm they fail.
Return a concise summary.

Input You Will Receive

Path to the plan file

Rules

Load test-writing-guide and code-writing-guide before writing or modifying any test code.
Tests must compile and run — but must fail at the end of this step. That is the expected outcome of Test First. Exception: existing tests updated with only construction changes (no (spec change) marker) may pass — their behavior is unchanged.
Do NOT implement any production code — test code only.
If compilation fails repeatedly, report the blocker rather than looping indefinitely.

Handling an Unexpected Pass

If tests pass when they should fail:

Unmarked updates to existing tests (no (spec change) marker) — only construction changed, behavior is unchanged; a pass is expected. Output STATUS: OK.
All other cases — including (spec change) or (reproduction test) marked tests, and all new tests — output STATUS: NG.

Output

Return a summary in this structure:

STATUS: OK  ← all tests failed as expected (or passes were expected per the rule above)
STATUS: NG  ← one or more tests passed unexpectedly

Place the STATUS: line first, then:

Which test files were added/modified
For STATUS: NG: list the tests that passed unexpectedly and why they were not judged legitimate

Changes

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.

3d ago First seen · 58 lines · 85 tokens per session scan A 46aeef2afe82

Subscribe to this mod's changes

failing-test-writer is an agent published in the GitHub repository nowsprinting/unity-coding-skills (19 stars, last pushed 3d ago), licensed Unlicense. It adds 85 tokens to every session and 561 once invoked, about $0.0004 per session on Opus 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 2026-08-30.

TDD Testing agent-skills claude-code-plugin coding-agent unity unity3d

implementer-agent

Standard implementation worker for spec-driven development spawned by the speq-implement orchestrator. Executes untagged tasks.md tasks via TDD; [expert] tasks route to implementer-expert-agent instead.

marconae/speq-skill · 45 tokens

uss-bem-checker

Reviews USS stylesheets and the C# strings that reference them against the Aspid.FastTools BEM grammar (class names) and the positional grammar (custom properties). Use after edits to any .uss file or to any code holding USS class names / --aspid- variables (Constants.cs, AspidStyles.cs, component .cs files).

VPDPersonal/Aspid.FastTools · 77 tokens

failing-test-writer