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

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

65%

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 as an overview, but its in-body guidance is abstract with all executable detail offloaded to references, and navigation is weakened by a broken path and several unreferenced bundle files. Validation exists only as a terminal step without a feedback loop.

Suggestions

Replace the vague "See Terraform implementation details" with an explicit link to ${CLAUDE_SKILL_DIR}/references/implementation.md, and fix the broken ${CLAUDE_SKILL_DIR}/examples/ reference to point to references/examples.md.

Add a concrete snippet or the init-terraform.sh invocation inline in the Instructions (e.g. the backend block or `bash scripts/init-terraform.sh <project>`) so the body is actionable without forcing a file hop.

Insert an explicit validate→fix→retry checkpoint after provisioning (e.g. run `terraform validate`/`plan` and only proceed on success) rather than a single terminal validation step.

Reference the remaining bundle files (ARD.md, PRD.md, scripts/init-terraform.sh) from the body so all provided materials are discoverable.

Dimension	Reasoning	Score
Conciseness	The body is lean — terse section headings, bullet prerequisites, and one-line instruction steps — and does not re-explain concepts Claude already knows (e.g. what Terraform or VPC-SC is).	3 / 3
Actionability	Steps name concrete targets (project_id, region, Private Service Connect, code execution, memory bank) but the body itself contains no executable Terraform or commands — all concrete code is deferred to referenced files, leaving guidance incomplete in place.	2 / 3
Workflow Clarity	An eight-step numbered sequence with a terminal "Validate Deployment" step is present, but there is no validate-then-fix feedback loop or explicit checkpoint gating for this batch infrastructure operation, capping the score at 2.	2 / 3
Progressive Disclosure	Some references are well-signaled (errors.md, examples.md), but "See Terraform implementation details" is vague, `${CLAUDE_SKILL_DIR}/examples/` points to a non-existent directory (examples live in references/examples.md), and several bundle files (implementation.md, ARD.md, PRD.md, init-terraform.sh) are never referenced from the body.	2 / 3
	Total	9 / 12 Passed

Description

100%

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 specific, complete, and distinctive, naming concrete infrastructure components and providing explicit natural-language trigger phrases. Minor awkward phrasing ("Execute use when") does not materially reduce quality.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities — "Provisions Agent Engine runtime, 14-day code execution sandbox, Memory Bank, VPC Service Controls, IAM roles, and secure multi-agent infrastructure" — matching the anchor for several specific concrete actions.	3 / 3
Completeness	Explicitly answers both what ("Provisions Agent Engine runtime...") and when ("Trigger with phrases like...") with explicit trigger guidance, satisfying the top anchor.	3 / 3
Trigger Term Quality	Provides good coverage of natural phrases a user would say ("deploy ADK terraform", "agent engine infrastructure", "provision ADK agent", "vertex AI agent terraform", "code execution sandbox terraform").	3 / 3
Distinctiveness Conflict Risk	Targets a clear niche — Vertex AI ADK infrastructure via Terraform — with distinct triggers unlikely to conflict with other 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
Commit: 3022dd3

Reviewed: 2 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.