CtrlK
BlogDocsLog inGet started
Tessl Logo

create-agent

Create and update Harness AI agent instances - standalone templates for agentic workflows in pipelines. Use when asked to create agent, update agent, modify agent spec, build autonomous systems, or work with AI agents.

88

6.84x
Quality

85%

Does it follow best practices?

Impact

89%

6.84x

Average score across 3 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

77%Weight 40%Scale 1-3

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

Highly actionable and clearly sequenced with strong validation checkpoints, but verbose: the same agent-config rules are repeated across phases and the table, and everything lives inline with no reference files for progressive disclosure.

Suggestions

De-duplicate the env-var/layout/UID rules: state them once (e.g. in Phase 3) and have the CRITICAL GUIDELINES table and Phase 2 reference rather than re-explain.

Extract the worked example and the v0/v1 pipeline usage into a reference file (e.g. examples.md) linked from a leaner overview to improve progressive disclosure.

Move the full default-configuration YAML reference into a separate reference doc and summarize only the essentials in SKILL.md.

DimensionReasoningScore

Conciseness

The body is largely useful and specific, but the default config is restated multiple times (Phase 2 default structure, Phase 3 assembly, the Example, and the CRITICAL GUIDELINES table repeat the same env-var/layout rules), adding tokens Claude already has in the same file.

2 / 3

Actionability

Provides copy-paste-ready YAML blocks, explicit env var values, MCP call signatures with parameters, and a fully worked Code Review Agent example plus v0/v1 pipeline usage.

3 / 3

Workflow Clarity

Five clearly sequenced phases with explicit validation checkpoints (interactive review-and-wait-for-confirmation before create/update, 'only after confirmation', check-existing-first rule), forming a complete feedback loop.

3 / 3

Progressive Disclosure

Content is well sectioned but entirely inline in a ~460-line SKILL.md with no bundle files; the full config reference, example, and pipeline usage could be split into one-level-deep references for a leaner overview.

2 / 3

Total

10

/

12

Passed

Description

92%Weight 40%Scale 1-3

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

A strong, third-person description with explicit capabilities and a clear 'Use when' trigger clause. Slight risk of over-broad triggers ('build autonomous systems', 'work with AI agents') diluting distinctiveness.

Suggestions

Tighten 'build autonomous systems' and 'work with AI agents' to Harness-specific phrasing (e.g. 'Use when asked to create or modify a Harness AI agent or agent spec') to reduce overlap with generic agentic skills.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('Create and update Harness AI agent instances', 'modify agent spec') plus the artifact ('standalone templates for agentic workflows in pipelines').

3 / 3

Completeness

Answers both what (create/update Harness AI agent instances/templates) and when via an explicit 'Use when asked to...' clause enumerating trigger scenarios.

3 / 3

Trigger Term Quality

Includes natural phrases users would say ('create agent', 'update agent', 'modify agent spec', 'build autonomous systems', 'work with AI agents') with good coverage of phrasings.

3 / 3

Distinctiveness Conflict Risk

The 'Harness AI agent' niche is reasonably distinct, but broad phrases like 'build autonomous systems' and 'work with AI agents' could overlap with other agentic/general-AI skills.

2 / 3

Total

11

/

12

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
harness/harness-ai
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.