adk-infra-expert

Execute use when provisioning Vertex AI ADK infrastructure with Terraform. Trigger with phrases like "deploy ADK terraform", "agent engine infrastructure", "provision ADK agent", "vertex AI agent terraform", or "code execution sandbox terraform". Provisions Agent Engine runtime, 14-day code execution sandbox, Memory Bank, VPC Service Controls, IAM roles, and secure multi-agent infrastructure.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

65%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.

The body is concise and well-structured with a logical workflow, but it stays at an abstract level: no executable Terraform/code, vague validation, and partial/broken bundle navigation hold it back. Tightening actionability and fixing reference paths would lift it substantially.

Suggestions

Add concrete, executable guidance to the Instructions: a minimal HCL snippet or commands (e.g. terraform init/plan/apply) and reference scripts/init-terraform.sh by path so steps are copy-paste ready.

Make validation explicit: name the concrete check (e.g. 'gcloud ai agent-engines describe <id>' or a curl against the agent endpoint) and add a validate-fix-retry loop rather than 'verify all components'.

Fix progressive-disclosure references: link 'Terraform implementation details' to references/implementation.md by path, correct the non-existent ${CLAUDE_SKILL_DIR}/examples/ reference, and surface ARD.md/PRD.md/init-terraform.sh where relevant.

Dimension	Reasoning	Score
Conciseness	The body is lean with a one-sentence overview, concise prerequisite bullets, and one-line steps; it avoids explaining concepts Claude already knows.	3 / 3
Actionability	Steps name concrete constructs (Private Service Connect, project_id, region, least-privilege roles) but are abstract one-liners with no executable Terraform or commands, and the existing init-terraform.sh is never referenced.	2 / 3
Workflow Clarity	The 8-step sequence is clearly ordered and includes a final validation step, but validation is vague ('Test agent endpoint and verify all components') with no concrete validation command or validate-fix-retry feedback loop for infrastructure provisioning.	2 / 3
Progressive Disclosure	One-level refs to errors.md and examples.md are well-signaled, but 'See Terraform implementation details' lacks a path (implementation.md exists), ${CLAUDE_SKILL_DIR}/examples/ does not exist, and ARD.md, PRD.md, and init-terraform.sh are never referenced.	2 / 3
	Total	9 / 12 Passed

Description

100%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.

The description is well-constructed: it uses third person, names concrete capabilities, and provides explicit trigger phrases covering what and when. Minor oddity is the opening 'Execute use when' phrasing, but all required elements are present.

Dimension	Reasoning	Score
Specificity	"Provisions Agent Engine runtime, 14-day code execution sandbox, Memory Bank, VPC Service Controls, IAM roles, and secure multi-agent infrastructure" lists multiple specific concrete capabilities in third person.	3 / 3
Completeness	It states both the what (the provisions list) and explicit when guidance ("Execute use when provisioning..." plus "Trigger with phrases like..."), satisfying the explicit-trigger requirement.	3 / 3
Trigger Term Quality	"Trigger with phrases like 'deploy ADK terraform', 'provision ADK agent', 'vertex AI agent terraform'" gives good coverage of natural terms a user would say for this domain.	3 / 3
Distinctiveness Conflict Risk	The Vertex AI ADK + Terraform provisioning niche with domain-specific triggers is clearly distinguishable and unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: jeremylongshore/claude-code-plugins-plus-skills
Path: plugins/devops/jeremy-adk-terraform/skills/adk-infra-expert/SKILL.md
Commit: 399d94c

Reviewed: 3 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.