CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

add-harness-package

Guide for adding new AI SDK harness packages. Use when creating a new @ai-sdk/harness-<name> package that adapts a coding-agent runtime to HarnessV1.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 content is a well-sequenced, highly actionable implementation guide with concrete commands, exact structure templates, explicit validation steps, and a final checklist. Minor room for improvement: trimming a few overlong asides and moving the dense bridge dependency rules into a reference file.

DimensionReasoningScore

Conciseness

The body is dense with project-specific requirements (file tree, package.json fields, bridge lockfile rules) and avoids explaining general concepts Claude already knows. A few long parenthetical asides (model-default handling, gateway base-URL conventions) could be trimmed, placing it at anchor 4 rather than 5.

4 / 5

Actionability

Provides copy-paste-ready commands (pnpm konsistent, pnpm update-references, pnpm --filter @ai-sdk/harness-<name> build/test, the exact bridge lockfile regeneration command), an exact baseline file structure, concrete package.json settings, and a concrete test-mocking pattern for the compiled-bridge ENOENT issue. This matches anchor 5: fully executable guidance covering the common cases.

5 / 5

Workflow Clarity

Ten clearly sequenced steps with explicit validation checkpoints (konsistent with 'fix any violations flagged before proceeding', the full build/test/type-check command block) and a feedback loop ('run relevant harness examples against a live sandbox early'), plus a comprehensive checklist. This matches anchor 5: explicit validation steps, error-recovery guidance, and a checklist for a complex process.

5 / 5

Progressive Disclosure

The single bundle reference (references/harness-requirements.md) exists, is clearly signaled at the top with its purpose, and is one level deep; the body itself is well-sectioned for a step-by-step guide. The detailed bridge dependency rules inlined in step 2 are dense enough that they could live in a separate reference file, keeping this at anchor 4 rather than 5.

4 / 5

Total

18

/

20

Passed

Description

73%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 third-person, concise, and clearly distinguishes this niche skill with an explicit 'Use when' trigger clause. Its main weakness is specificity: it names the deliverable but does not enumerate the concrete actions the guide covers, leaving the 'what' at a fairly high level.

Suggestions

Enumerate 2-3 concrete actions the guide covers (e.g., scaffolding the package structure, configuring bridge dependencies, adding tests and docs) to raise specificity.

Add a natural trigger variation or two (e.g., 'adding a harness adapter for a new coding-agent runtime') to broaden trigger-term coverage without adding jargon.

DimensionReasoningScore

Specificity

Names the domain and a concrete deliverable ("creating a new @ai-sdk/harness-<name> package that adapts a coding-agent runtime to HarnessV1") but lists only 1-2 actions rather than several specific capabilities. It is below anchor 4 because the coverage of what the guide entails is not enumerated, and above anchor 2 because the actions named are concrete, not generic.

3 / 5

Completeness

Explicitly answers both: what ("Guide for adding new AI SDK harness packages") and when ("Use when creating a new @ai-sdk/harness-<name> package..."). The 'when' clause is explicit but consists of a single condition without broader trigger phrasing, matching anchor 4 rather than 5; it is clearly above anchor 3 because 'when' is not merely implied.

4 / 5

Trigger Term Quality

Includes natural phrases a user would say ("adding new AI SDK harness packages", "creating a new @ai-sdk/harness-<name> package") plus "coding-agent runtime". Good keyword coverage but a few natural terms/synonyms are missing and "HarnessV1" is technical jargon, so it falls just below anchor 5.

4 / 5

Distinctiveness Conflict Risk

Scoped to a highly specific artifact (@ai-sdk/harness-<name> packages adapting runtimes to HarnessV1), giving it a clear niche with distinct triggers and minimal conflict risk with any other skill, matching anchor 5 exactly.

5 / 5

Total

16

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

relative_links

Relative link issues: 2 suspicious

Warning

Total

13

/

16

Passed

Repository

vercel/ai

Path

skills/add-harness-package/SKILL.md

Reviewed

Table of Contents

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.