create-environments

Create or migrate native verifiers.v1 taskset, environment, and harness packages. Use to build a taskset, port a benchmark, add task tools, script or model a user, build a multi-agent environment, package an agent harness, or migrate an existing v0 environment to the typed v1 trace model.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

82%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a high-quality, actionable authoring guide: executable code, concrete commands, explicit validation checkpoints, and a clear publish gate. It is held back from the top band mainly by run-on paragraphs that hurt conciseness and a topical rather than strictly sequenced organization.

Suggestions

Break the long run-on paragraphs in "User simulation" and "Multi-agent environments" into short bullet points or labeled sub-bullets so each rule earns its place and is easier to follow.

Add a short numbered end-to-end workflow (bootstrap → define values → implement taskset → add env/harness if needed → validate → publish gate) near the top, with the topical sections as reference depth beneath it.

Consider moving the detailed ownership/API-contract rules into a references file (e.g. references/contract.md) and keeping SKILL.md as a leaner overview, so progressive disclosure can reach the top band.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence — no padding about basic concepts — with terse imperative style and complete code. Not 5 because a few sections (e.g. the "Multi-agent environments" and "User simulation" paragraphs) are long run-on sentences that could be tightened or split for readability.	4 / 5
Actionability	Provides copy-paste-ready, executable code (full AdditionTask/AdditionTaskset example, SearchToolset), concrete CLI commands ("uv run init", "prime images push", "prime env push"), and named API surfaces covering the common cases. Not 4 because the examples are complete and runnable rather than having gaps.	5 / 5
Workflow Clarity	Contains a planning checklist ("Define the needed values first"), an explicit lifecycle sequence (setup → harness execution → finalize → reward/metric), and a validated publish gate ("Only then run ..."), so validation checkpoints are present and the destructive-operation cap does not apply. Not 5 because the document is organized topically rather than as one end-to-end numbered workflow with feedback loops throughout.	4 / 5
Progressive Disclosure	Well-organized into clearly headed sections with one-level-deep, signaled doc references ("also see docs/v1/harbor.md", "See docs/v1/env.md") and no nested reference chains. Not 5 because substantial reference-style material (ownership rules, API contracts) is inlined in SKILL.md and no bundle files exist to offload it, so it is not a lean overview pointing to split detail files.	4 / 5
	Total	17 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it states a concrete what, gives explicit "Use to ..." trigger guidance with many specific actions, and carves out a distinct niche. Its only weakness is a technical-jargon bias and missing synonyms/file extensions in the trigger terms.

Dimension	Reasoning	Score
Specificity	Lists seven concrete actions — "build a taskset, port a benchmark, add task tools, script or model a user, build a multi-agent environment, package an agent harness, or migrate an existing v0 environment" — alongside the opening "Create or migrate ... taskset, environment, and harness packages", giving comprehensive concrete coverage. Not 4 because the action list is broad and specific rather than having only minor gaps.	5 / 5
Completeness	Explicitly answers what ("Create or migrate native verifiers.v1 taskset, environment, and harness packages") and when via the "Use to ..." clause with concrete trigger phrases, satisfying the anchor for both. Not 4 because the when-guidance is explicit and concrete, not merely implied.	5 / 5
Trigger Term Quality	Good natural trigger coverage ("build a taskset", "port a benchmark", "build a multi-agent environment", "package an agent harness") but leans technical with jargon like "verifiers.v1", "typed v1 trace model", and "harness", and lacks synonyms or file extensions. Not 5 because common-phrase/synonym coverage is incomplete; not 3 because several genuine user-sayable phrases are present.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (verifiers.v1 native package authoring and v0→v1 migration) with distinct, framework-specific triggers and minimal overlap risk with other skills. Not 4 because the domain scoping is tight enough that confusion with unrelated skills is unlikely.	5 / 5
	Total	19 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: PrimeIntellect-ai/verifiers
Path: skills/create-environments/SKILL.md
Commit: c51c094

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.