What is ensure-test-coverage?

A testing skill for one evaluation: it reviews existing tests, finds parts that need coverage, and can create or improve tests. An evaluation is a defined task used to measure how well a system performs.

Use it to inspect tasks, solvers, scorers, tools, and dataset handling for a named evaluation, then report or fill its testing gaps.

It helps reveal untested code and checks that tests follow the repository's conventions, so important evaluation behaviour is less likely to go unchecked.

ensure-test-coverage — Skill for Claude Code by UKGovernmentBEIS

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo

git clone --depth 1 https://github.com/UKGovernmentBEIS/inspect_evals

agentmods

npx agentmods add skills/ukgovernmentbeis/inspect_evals/ensure-test-coverage

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/ukgovernmentbeis/inspect_evals/ensure-test-coverage/github.svg)](https://agentmods.dev/skills/ukgovernmentbeis/inspect_evals/ensure-test-coverage)

Your own site

<a href="https://agentmods.dev/skills/ukgovernmentbeis/inspect_evals/ensure-test-coverage"><img src="https://agentmods.dev/badge/skills/ukgovernmentbeis/inspect_evals/ensure-test-coverage/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/ukgovernmentbeis/inspect_evals/ensure-test-coverage"><img src="https://agentmods.dev/badge/skills/ukgovernmentbeis/inspect_evals/ensure-test-coverage.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 105 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,665 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

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 5.1	$0.00105	$0.04665
Opus 5	$0.00053	$0.02332
Sonnet 5	$0.00021	$0.00933
Haiku 4.5	$0.00011	$0.00466

Measured today against content hash 1e27f0404ad1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

ensure-test-coverage 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 today.

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.

.claude/skills/ensure-test-coverage/SKILL.md · 352 lines

How it starts

The opening of the file, as written. The whole thing — 352 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ensure Test Coverage

Analyze a single evaluation's test coverage against repository testing conventions. This skill can both review existing tests and create missing ones.

Expected Arguments

When invoked, this skill expects:

eval_name (required): The name of one or more evaluations to check (e.g., gpqa, healthbench)

If not provided, ask the user.

Workflow

Phase 1: Discover Testable Components

Analyze the eval's source code in src/inspect_evals/<eval_name>/ to build a complete inventory of what needs testing.

1.1 Scan for decorated functions

Use AST analysis or grep to find all functions decorated with:

@task - Task entry points (and their parameters/variants)
@solver - Custom solvers
@scorer - Custom scorers
@tool - Custom tools

For each, record: function name, file path, line number, parameters.

1.2 Scan for dataset patterns

Look for:

record_to_sample functions (named exactly or as a lambda/callable passed to hf_dataset/csv_dataset/json_dataset)
hf_dataset() or load_dataset() calls (indicates HuggingFace datasets)
csv_dataset() or json_dataset() calls
Dataset URLs or paths (for understanding data sources)

1.3 Scan for sandbox usage

Look for:

compose.yaml or docker-compose.yaml files in the eval's data/ directory
sandbox() calls in source code
SandboxEnvironmentSpec references

1.4 Scan for utility functions

Look for non-trivial helper functions that contain meaningful logic (not just thin wrappers). Functions with:

Complex branching (if/elif chains)
Data transformations
String parsing/formatting
Score calculations
File I/O operations

1.5 Identify task variants

Check if the eval has multiple @task functions or a single task with parameters that create meaningfully different variants (e.g., different subsets, different Docker environments, different scorers).

Phase 2: Run Autolint and Measure Coverage

2.1 Run autolint

Read the full file on GitHub · 352 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

references/test-patterns.md 13 KB

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.

today Changed · +11 lines 1e27f0404ad1
10d ago First seen · 341 lines · 105 tokens per session scan A 3011d19be6e4

Subscribe to this mod's changes

ensure-test-coverage is a skill published in the GitHub repository UKGovernmentBEIS/inspect_evals (664 stars, last pushed today), licensed MIT. It adds 105 tokens to every session and 4,665 once invoked, about $0.0005 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.

Testing

research-engineer

An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.

davila7/claude-code-templates · 43 tokens

tika-eval-compare

Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".

apache/tika · 50 tokens

neuron-evaluation-engineer

Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…

neuron-core/neuron-ai · 77 tokens

ensure-test-coverage