hosted-agents

tessl i github:muratcankoylan/Agent-Skills-for-Context-Engineering --skill hosted-agents

This skill should be used when the user asks to "build background agent", "create hosted coding agent", "set up sandboxed execution", "implement multiplayer agent", or mentions background agents, sandboxed VMs, agent infrastructure, Modal sandboxes, self-spawning agents, or remote coding environments.

57%

Overall

Validation — 75%

Implementation — 57%

Activation — 37%

SKILL.md

Review

Evals

Validation

75%

Warnings & errors only

Criteria	Description	Result
metadata_version	'metadata' field is not a dictionary	Warning
license_field	'license' field is missing	Warning
body_examples	No examples detected (no code fences and no 'Example' wording)	Warning
body_output_format	No obvious output/return/format terms detected; consider specifying expected outputs	Warning

	Total	12 / 16 Passed

Implementation

57%

This skill provides comprehensive architectural guidance for hosted agent infrastructure with good organization and clear topic separation. However, it reads more like a design document than an actionable skill - it describes patterns and approaches without providing executable code or concrete implementation steps. The content would benefit significantly from code examples for key operations like sandbox setup, image building, and API endpoints.

Suggestions

Add executable code examples for critical operations: sandbox image building script, snapshot/restore commands, and API endpoint implementations

Include validation checkpoints in multi-step workflows, especially for image building and session state management (e.g., 'Verify image health before adding to pool')

Remove or condense the 'Core Concepts' section - Claude understands why hosted agents are useful; focus on the how

Add concrete configuration examples for the recommended tools (Modal sandbox config, Durable Objects setup, GitHub App manifest)

Dimension	Reasoning	Score
Conciseness	The skill contains some unnecessary conceptual explanations (e.g., 'Core Concepts' section explaining why hosted agents matter) that Claude already understands. However, most content is practical and relevant, with good density of actionable information.	2 / 3
Actionability	Provides good conceptual guidance and patterns but lacks executable code examples. Descriptions like 'Pre-build environment images' and 'Take filesystem snapshots' are directional rather than copy-paste ready. No concrete code for sandbox setup, API implementation, or client integration.	2 / 3
Workflow Clarity	The 'Sandbox-to-API Flow' section shows a clear sequence, but most multi-step processes lack explicit validation checkpoints. Critical operations like image building, snapshot management, and session state handling don't include error recovery or verification steps.	2 / 3
Progressive Disclosure	Well-organized with clear sections, appropriate use of headers, and references to related skills and external resources. The structure allows scanning for relevant topics. References are one level deep and clearly signaled.	3 / 3
	Total	9 / 12 Passed

Activation

37%

This description is fundamentally incomplete - it functions purely as a trigger list without explaining what the skill actually does. While the trigger terms are well-chosen and specific, the complete absence of capability descriptions makes it impossible for Claude to understand what actions this skill enables or how it differs from other agent-related skills.

Suggestions

Add a capability statement at the beginning describing what the skill does (e.g., 'Builds and deploys background coding agents in sandboxed Modal VMs with self-spawning capabilities and multiplayer collaboration support.')

Restructure to follow the pattern: '[What it does]. Use when [triggers].' to ensure both components are present

Include specific concrete actions like 'configures VM environments', 'sets up agent orchestration', or 'implements spawn logic' to clarify the skill's actual capabilities

Dimension	Reasoning	Score
Specificity	The description contains no concrete actions - it only lists trigger phrases without explaining what the skill actually does. There are no verbs describing capabilities like 'creates', 'configures', or 'deploys'.	1 / 3
Completeness	The description only addresses 'when' (trigger conditions) but completely omits 'what' - there is no explanation of what capabilities or actions this skill provides. The 'what' component is entirely missing.	1 / 3
Trigger Term Quality	Excellent coverage of natural trigger terms users would say: 'build background agent', 'create hosted coding agent', 'set up sandboxed execution', 'Modal sandboxes', 'remote coding environments'. These are specific and varied.	3 / 3
Distinctiveness Conflict Risk	The trigger terms are fairly specific to background/sandboxed agent infrastructure, but without knowing what the skill does, there's potential overlap with general agent-building or infrastructure skills. Terms like 'agent infrastructure' are somewhat broad.	2 / 3
	Total	7 / 12 Passed

Reviewed

16 days ago

Table of Contents

Validation Implementation Activation

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.