docker-sandbox

Create, manage, and execute agent tools (claude, codex) inside Docker sandboxes for isolated code execution. Use when running agent loops, spawning tool subprocesses, or any task requiring process isolation. Triggers on "sandbox", "isolated execution", "docker sandbox", "safe agent execution", or when working on agent loop infrastructure.

Quality

70%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/docker-sandbox/SKILL.md

Quality

Content

57%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.

The body is a concise, competent set of guardrails for sandbox operation but reads as unstructured prose heavy on policy and light on concrete executable steps. Adding headers, an explicit ordered workflow, and clearer pointers to sibling skills would materially improve it.

Suggestions

Add section headers (e.g., ## Resolve request type, ## Authentication, ## Isolation, ## Ports & cleanup) so the workflow is scannable instead of a wall of prose.

Provide concrete executable commands or an explicit numbered sequence with a validate->fix->retry checkpoint for destructive operations (container start/stop, port publishing).

Surface references to sandbox-computer and workflow-rig as clearly signaled pointers (e.g., "See the `sandbox-computer` skill") rather than burying them mid-paragraph.

Dimension	Reasoning	Score
Conciseness	Lean prose that assumes Claude's competence and avoids explaining Docker or sandboxing basics; each paragraph earns its place, with only minor tightening possible.	4 / 5
Actionability	Gives some concrete directives (resolve demo vs runtime, use sandbox-computer/workflow-rig, inspect directory, preserve files, publish only requested ports) but is mostly policy and prohibitions with few specific executable commands, deferring specifics to "live help for the selected tool".	3 / 5
Workflow Clarity	A coherent implicit sequence exists with verification checkpoints ("verifying the requested workspace", "Verify the expected service before reporting a URL", "Report... checks"), but it is prose rather than an explicit ordered workflow and operates on destructive/batch operations without a full validate->fix->retry loop.	3 / 5
Progressive Disclosure	The body is short and references sibling skills (sandbox-computer, workflow-rig) inline, but it lacks section headers, reads as a wall of prose, and the references are not clearly signaled as navigable pointers.	3 / 5
	Total	13 / 20 Passed

Description

83%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.

The description is strong: it clearly states capabilities, gives explicit "Use when" and "Triggers on" guidance, and occupies a distinct niche. Minor improvements would be replacing the generic "manage" with concrete verbs and adding a couple more trigger synonyms.

Suggestions

Replace the generic verb "manage" with specific concrete actions (e.g., "provision, configure, and tear down") to lift specificity.

Add a few more natural trigger synonyms users might say (e.g., "containerized agent", "isolated tool runtime") to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ("Create, manage, and execute agent tools (claude, codex) inside Docker sandboxes"), but "manage" is generic and coverage of what those actions entail is not fully comprehensive.	4 / 5
Completeness	Explicitly answers both what (create/manage/execute agent tools in Docker sandboxes) and when ("Use when running agent loops, spawning tool subprocesses..." plus an explicit "Triggers on" clause with concrete trigger phrases).	5 / 5
Trigger Term Quality	Provides good natural trigger coverage ("sandbox", "isolated execution", "docker sandbox", "safe agent execution") a user would actually say, but is missing a few common synonyms/variants.	4 / 5
Distinctiveness Conflict Risk	Targets a clear niche (agent tools claude/codex inside Docker sandboxes) with distinct triggers, with only minor overlap risk against general Docker or sandbox skills.	4 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: joelhooks/joelclaw
Path: skills/docker-sandbox/SKILL.md
Commit: 7094ced

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.