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

71%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/ai-ml/jeremy-adk-orchestrator/skills/adk-deployment-specialist/SKILL.md

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 lean and well-sequenced with explicit validation checkpoints in the deploy flow, but it stays at a procedural level without inline executable examples and leaves several bundled reference files unlinked, weakening actionability and navigation.

Suggestions

Add one copy-paste-ready snippet for the core deployment (e.g., a minimal `vertexai.Client.agent_engines.create(...)` call with the required parameters) so the deploy step is executable inline rather than only described.

Add an explicit feedback loop after the deploy/validate step (e.g., 'If endpoint validation fails → review logs, fix the agent, redeploy, re-validate') to close the workflow for a risky deployment operation.

Link or remove the orphaned bundle files (ARD.md, PRD.md, how-it-works.md, workflow-examples.md) and fill in the empty 'ADK docs:' resource so all references are reachable and navigation is complete.

Dimension	Reasoning	Score
Conciseness	The ~40-line body assumes Claude's competence, avoids explaining concepts like what ADK or A2A are, and each section earns its place; the only slack is a dangling 'ADK docs:' bullet, but overall every token pulls weight.	3 / 3
Actionability	Steps such as "Define the AgentCard + A2A interfaces" and "Implement the agent(s)" are concrete and domain-specific but lack inline executable code or exact commands; the lone `vertexai.Client.agent_engines.create()` mention is described rather than shown, leaving key deploy/test details in referenced files.	2 / 3
Workflow Clarity	A clear six-step sequence is given and steps 4–5 name validation ("add smoke tests", "validate the generated endpoints"), but for a deployment workflow the fail→fix→redeploy feedback loop is only implicit, which the rubric caps at 2 for destructive/risky operations.	2 / 3
Progressive Disclosure	SKILL.md is a concise overview with one-level-deep references to errors.md and examples.md (both present), but several bundle files (ARD.md, PRD.md, how-it-works.md, workflow-examples.md) are never referenced from the body and the 'ADK docs:' link is empty, so navigation is incomplete.	2 / 3
	Total	9 / 12 Passed

Description

77%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 clearly conveys both capability and trigger conditions with strong specificity, but its trigger terms mix targeted phrases with generic ones ('infrastructure', 'CI/CD') that raise conflict risk with adjacent deployment skills.

Suggestions

Tighten trigger terms to ADK-specific phrasing (e.g., 'deploy ADK agent to Vertex AI', 'orchestrate A2A agents') and drop or qualify the generic 'infrastructure' trigger to reduce overlap with general deployment skills.

Add a couple of natural-language variants users would actually say (e.g., 'set up Agent Engine', 'wire up A2A agents') to improve trigger coverage beyond the current technical terms.

Dimension	Reasoning	Score
Specificity	"Deploy and orchestrate Vertex AI ADK agents" and "Manages AgentCard discovery, task submission, Code Execution Sandbox, and Memory Bank" list multiple concrete actions within a defined domain, matching the anchor that rewards several specific actions.	3 / 3
Completeness	It explicitly states both what ("Deploy and orchestrate Vertex AI ADK agents using A2A protocol") and when ("Use when asked to 'deploy ADK agent' or 'orchestrate agents'" plus explicit trigger phrases), satisfying the what-and-when anchor.	3 / 3
Trigger Term Quality	"Use when asked to 'deploy ADK agent'" and "Trigger with phrases like 'deploy', 'infrastructure', or 'CI/CD'" include relevant keywords, but 'infrastructure' is generic and several natural variants are framed as technical jargon rather than conversational triggers, so coverage is partial.	2 / 3
Distinctiveness Conflict Risk	The ADK/A2A/Vertex AI qualifiers carve a fairly distinct niche, but the standalone trigger terms 'deploy', 'infrastructure', and 'CI/CD' are generic and could overlap with a general deployment or CI skill.	2 / 3
	Total	10 / 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/ai-ml/jeremy-adk-orchestrator/skills/adk-deployment-specialist/SKILL.md
Commit: 399d94c

Reviewed: 1 day 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.