adk-deployment-specialist

Deploy and orchestrate Vertex AI ADK agents using A2A protocol. Manages AgentCard discovery, task submission, Code Execution Sandbox, and Memory Bank. Use when asked to "deploy ADK agent" or "orchestrate agents". Trigger with phrases like 'deploy', 'infrastructure', or 'CI/CD'.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

77%

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

A well-structured, concise overview with a clear validated workflow, but the body lacks executable code (deferring it to references) and its progressive disclosure is undermined by orphaned bundle files and a broken Resources section. Strengthening the in-body code examples and fixing navigation would raise it.

Suggestions

Add one minimal executable code block to the Instructions (e.g. a copy-paste `vertexai.Client.agent_engines.create(...)` deployment snippet) so the body is actionable without forcing a jump to examples.md.

Link all bundle files from the body — add an index pointing to how-it-works.md, workflow-examples.md, ARD.md, and PRD.md — so reference material is discoverable rather than orphaned.

Fix the Resources section: complete the empty 'ADK docs:' URL and remove or correct the dangling '000-docs/6767-a-SPEC-DR-STND-claude-code-plugins-standard.md' reference that does not exist in the bundle.

Dimension	Reasoning	Score
Conciseness	Lean body that assumes Claude's competence — no preamble explaining what ADK/A2A/Agent Engine are; sections are short and each line carries information rather than padding.	3 / 3
Actionability	Names concrete anchors (`/.well-known/agent-card`, `vertexai.Client.agent_engines.create()`, task send/status APIs) but the Instructions are abstract procedural steps with no executable code blocks in the body; copy-paste-ready code is deferred to examples.md, leaving key details missing here.	2 / 3
Workflow Clarity	A clear 6-step sequence with validation checkpoints — 'Test locally with representative prompts and failure cases', 'add smoke tests', 'validate the generated endpoints', plus retry/backoff observability and a verification checklist in Output — satisfying explicit validation and error-recovery anchors.	3 / 3
Progressive Disclosure	The two referenced files (errors.md, examples.md) are well-signaled one level deep, but navigation is incomplete: 4 of 6 bundle files (ARD.md, PRD.md, how-it-works.md, workflow-examples.md) are orphaned from the body, the 'ADK docs:' line is empty, and it references a non-existent path '000-docs/6767-a-SPEC-DR-STND-claude-code-plugins-standard.md'.	2 / 3
	Total	10 / 12 Passed

Description

92%

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 description: third-person, concrete, with an explicit 'Use when' trigger clause covering what and when. Its only weakness is the broad single-word triggers ('infrastructure', 'CI/CD') that raise conflict risk with generic deployment skills.

Suggestions

Narrow the trigger terms: replace generic 'infrastructure'/'CI/CD' with ADK-specific phrasings (e.g. 'deploy to Agent Engine', 'Vertex AI agent', 'A2A AgentCard') to reduce conflict with general DevOps skills.

Drop the redundant 'Trigger with phrases like...' sentence since the 'Use when asked to "deploy ADK agent" or "orchestrate agents"' clause already covers triggers.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'AgentCard discovery, task submission, Code Execution Sandbox, and Memory Bank' plus deploy/orchestrate — matching the 'lists multiple specific concrete actions' anchor.	3 / 3
Completeness	Explicitly answers both what ('Deploy and orchestrate Vertex AI ADK agents...Manages AgentCard discovery...') and when ('Use when asked to "deploy ADK agent" or "orchestrate agents"'), with an explicit 'Use when' trigger clause.	3 / 3
Trigger Term Quality	Provides natural phrasings users would say — 'deploy ADK agent' and 'orchestrate agents' — plus 'deploy', 'infrastructure', 'CI/CD', giving good coverage of likely user terms.	3 / 3
Distinctiveness Conflict Risk	The ADK/A2A/Agent Engine niche is distinct, but generic trigger words like 'infrastructure' and 'CI/CD' could fire for unrelated DevOps skills, creating overlap risk; not a clear 3.	2 / 3
	Total	11 / 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: 2cafb23

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.