architecture

Use when navigating the codebase for the first time, adding a new client method, adding a new container handler/service, or understanding how a request flows from Worker through the Sandbox DO into the container. Covers the three-layer architecture, client pattern, container runtime structure, and monorepo layout. (project)

1.97x

Quality

81%

Does it follow best practices?

Impact

95%

1.97x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

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

A well-structured, information-dense architecture reference with concrete paths and a sequenced workflow including a testing checkpoint. It could be tightened by removing the duplicated two-path explanation and improved by splitting deeper detail into referenced files.

Suggestions

Consolidate the 'Request Flow' ASCII diagrams with the 'Primary Control Path' and 'Route-Based Compatibility Path' sections to remove the duplicated explanation of the same two paths.

Extract detailed client/handler/service inventories (e.g., the specialized clients list and container runtime component breakdown) into a one-level-deep reference file, leaving SKILL.md as a tighter overview.

Add an explicit validation feedback loop to the 'adding a new container control operation' workflow (e.g., 'if tests fail, fix and re-run before mirroring the call').

Dimension	Reasoning	Score
Conciseness	The body is dense and information-rich, assuming Claude's knowledge (no padding about what Durable Objects or npm are), but the two-path structure is presented once as ASCII flow diagrams and then re-explained in the 'Primary Control Path' and 'Route-Based Compatibility Path' sections — minor redundancy that could be trimmed, stopping short of anchor 5.	4 / 5
Actionability	Provides concrete guidance with specific file paths ('packages/sandbox/src/container-control/') and a numbered procedure for adding a container control operation, but as an instruction-only architecture skill it offers no executable code or commands, placing it at anchor 4 rather than the copy-paste-ready anchor 5.	4 / 5
Workflow Clarity	The 'When adding a new container control operation' list is a clear 4-step sequence with a validation checkpoint ('Add unit tests on both sides; add an E2E test if it touches real shell/filesystem behavior'), but there is no explicit validate→fix→retry feedback loop, so it sits below anchor 5.	4 / 5
Progressive Disclosure	Content is well-organized into clearly headed sections (Three-Layer Architecture, Request Flow, Container Runtime, etc.) in a single file with no bundle references, but the document is longer than the ~50-line simple-skill threshold and is monolithic rather than split across one-level-deep reference files, keeping it at anchor 4.	4 / 5
	Total	16 / 20 Passed

Description

87%Weight 40%Scale 1-5

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, specific project-scoped description that clearly states both capabilities and trigger conditions with low conflict risk. Minor room to broaden capability phrasing and add synonym-style trigger variants.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — 'navigating the codebase for the first time, adding a new client method, adding a new container handler/service, ... understanding how a request flows from Worker through the Sandbox DO into the container' — but these read more as trigger tasks than enumerated capabilities, leaving minor coverage gaps versus the anchor 5 comprehensive list.	4 / 5
Completeness	Explicitly answers both what ('Covers the three-layer architecture, client pattern, container runtime structure, and monorepo layout') and when ('Use when navigating the codebase for the first time, adding a new client method...') with concrete trigger phrases, matching the anchor 5 example.	5 / 5
Trigger Term Quality	Natural developer phrases like 'navigating the codebase', 'adding a new client method', and 'adding a new container handler/service' match what a user would say, but it lacks synonyms or variants that would push it to anchor 5.	4 / 5
Distinctiveness Conflict Risk	The description is scoped to a specific project niche — 'request flows from Worker through the Sandbox DO into the container' and 'three-layer architecture' — with distinct triggers and a '(project)' tag, giving minimal conflict risk per anchor 5.	5 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: cloudflare/sandbox-sdk
Path: .agents/skills/architecture/SKILL.md
Commit: 6abf639

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.