openshell-cli

Guide agents through using the OpenShell CLI (openshell) for sandbox management, gateway registration, provider configuration and refresh, policy iteration, settings, service exposure, BYOC workflows, and inference routing. Covers basic through advanced multi-step workflows. Trigger keywords - openshell, sandbox create, sandbox exec, sandbox connect, logs, provider create, provider profile, provider refresh, policy set, policy get, settings, service expose, forward, port forward, BYOC, bring your own container, inference, use openshell, run openshell, CLI usage, manage sandbox, manage provider, gateway add, gateway select.

Quality

86%

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

Quality

Content

72%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 content is highly actionable and the workflows are exemplary with validation checkpoints and feedback loops. However it is far too verbose for a SKILL.md, inlining deep reference material that should live in the referenced cli-reference.md, which undermines progressive disclosure.

Suggestions

Move the dense behavioral/edge-case notes (OAuth RECOVERY codes, credential-backend migration, middleware WebSocket coverage, process.run_as_user UID rules) into cli-reference.md and keep SKILL.md to overviews plus the core command examples.

Trim explanatory prose that the CLI's own --help already provides; the skill states it is authoritative, so rely on it rather than restating flag semantics.

Reduce the Quick Reference table or relocate it to the reference file so the SKILL.md body stays a navigable overview rather than a monolithic manual.

Dimension	Reasoning	Score
Conciseness	At ~800 lines the body extensively explains gateway internals, OAuth recovery codes, credential-backend migration, and middleware WebSocket bindings that Claude could instead read via --help; substantial padding beyond what a CLI skill needs.	2 / 5
Actionability	Nearly every section gives copy-paste-ready bash commands with concrete flags and named arguments, plus a full Quick Reference table covering the common cases.	5 / 5
Workflow Clarity	Multi-step workflows (notably the Policy Iteration Loop) are explicitly sequenced with an ASCII diagram, validation via `policy list`/revision status, `--wait` exit codes, and explicit feedback loops for error recovery.	5 / 5
Progressive Disclosure	Only one external reference exists (cli-reference.md), yet the body still inlines exhaustive command details, flag tables, and behavioral notes that belong in that reference; the split is inverted from the intended overview-plus-reference structure.	2 / 5
	Total	14 / 20 Passed

Description

100%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 comprehensive and specific, clearly stating both the capability set and explicit trigger keywords. It is detailed but earns its length through concrete command coverage. Voice is appropriately third person ('Guide agents through...').

Suggestions

The keyword list is long and reads as keyword stuffing; trimming to the highest-signal natural phrases would preserve trigger quality while reducing token cost.

Dimension	Reasoning	Score
Specificity	Lists many concrete actions (sandbox create, sandbox exec, provider refresh, policy set/get, service expose, BYOC, inference routing), giving comprehensive coverage of the CLI's surface.	5 / 5
Completeness	Explicitly states what it does ('Guide agents through using the OpenShell CLI...') and includes explicit trigger guidance via the 'Trigger keywords' list, answering both what and when.	5 / 5
Trigger Term Quality	Includes an exhaustive keyword list spanning natural phrases ('use openshell', 'run openshell', 'manage sandbox'), CLI subcommands, and acronyms ('BYOC', 'bring your own container') that users would actually say.	5 / 5
Distinctiveness Conflict Risk	The OpenShell-specific command names and BYOC/inference niche make it clearly distinguishable from generic skills, with minimal overlap risk against other specialized skills.	5 / 5
	Total	20 / 20 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
skill_md_line_count	SKILL.md is long (819 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: NVIDIA/OpenShell
Path: .agents/skills/openshell-cli/SKILL.md
Commit: 69a05eb

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.