CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

onecli-gateway

OneCLI Gateway: transparent HTTPS proxy that injects stored credentials into outbound calls. You MUST use this skill when the user asks you to read emails, check calendar, access GitHub repos, create issues, check Stripe payments, or interact with ANY external service or API. Do NOT use browser extensions or OAuth CLI tools. Make HTTP requests directly; the gateway injects credentials automatically.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

81%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 lean, actionable, well-structured guide: executable curl examples, a concrete stub-creation recipe, and an explicitly sequenced failure-recovery workflow with retry/escalation loops. The only minor issues are repeated statements of 'the gateway injects credentials automatically' and the absence of a sample error response body.

DimensionReasoningScore

Conciseness

Sections are lean with copy-paste curl examples and no explanation of concepts Claude already knows, but 'the gateway injects credentials automatically' is repeated roughly four times across sections and could be trimmed. This matches anchor 4 (efficient with minor over-explanation) rather than anchor 5's 'every token earns its place'.

4 / 5

Actionability

The curl examples are executable and copy-paste ready, the stub recipe is concrete ('"onecli-managed" as the placeholder for all secret values, with file permissions 0600'), and the failure flow uses concrete fields ('connect_url from the error response'). It falls short of anchor 5 because no example error/response body is shown and the connect-link step is a template rather than an executable command.

4 / 5

Workflow Clarity

The failure workflow is explicitly sequenced with a genuine feedback loop: Step 1 present the connect_url, Step 2 retry after the user confirms, and 'If the retry still fails, ask if they need help' provides error-recovery escalation. The happy path is a single unambiguous action (call the real API URL directly), and there are no destructive or batch operations requiring additional validation, so anchor 5 applies.

5 / 5

Progressive Disclosure

The body is well organized into clear sections with no nested references and a single clearly-signaled external doc URL for credential stubs. At roughly 70 lines it sits just above the under-50-line threshold that would allow a 5 without external references, so anchor 4 (good structure, minor organization considerations) is the better fit.

4 / 5

Total

17

/

20

Passed

Description

71%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 clearly and explicitly states what the skill does and when to use it, with natural trigger phrases for common services. Its weaknesses are the second-person voice (penalized on specificity) and the over-broad 'ANY external service or API' trigger, which creates substantial conflict risk with other skills.

Suggestions

Rewrite in third person ('Injects stored credentials into outbound HTTPS calls...') to avoid the second-person specificity penalty.

Replace 'interact with ANY external service or API' with a bounded list of concrete triggers (e.g., Gmail, Google Calendar, GitHub, Stripe) to reduce conflict with other network-related skills.

Add a few common synonyms and file/service extensions users would naturally say (e.g., Gmail, send email, calendar events, pull requests) to improve trigger term coverage.

DimensionReasoningScore

Specificity

The description names concrete actions and services ('read emails, check calendar, access GitHub repos, create issues, check Stripe payments', 'Make HTTP requests directly'), which matches the anchor-4 'several specific actions' level, but the rubric's second-person penalty ('You MUST use this skill...') reduces it by 1. It is not anchor 2 because the domain and actions are specific rather than generic.

3 / 5

Completeness

It explicitly answers both 'what' ('transparent HTTPS proxy that injects stored credentials into outbound calls') and 'when' ('You MUST use this skill when the user asks you to read emails, check calendar, access GitHub repos...') with concrete trigger phrases, matching the anchor-5 example structure.

5 / 5

Trigger Term Quality

Natural user phrasings are present ('read emails', 'check calendar', 'access GitHub repos', 'create issues', 'check Stripe payments'), giving good keyword coverage. It falls short of anchor 5 because common synonyms and variations (e.g., Gmail, send email, calendar events) are missing.

4 / 5

Distinctiveness Conflict Risk

The trigger 'interact with ANY external service or API' is maximally broad and would cause this skill to fire for nearly any network-related task, creating high overlap risk with many other skills. The 'Do NOT use browser extensions or OAuth CLI tools' carve-outs only partially mitigate this, so it fits anchor 2 rather than anchor 3's 'somewhat specific' overlap.

2 / 5

Total

14

/

20

Passed

Validation

100%

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

nanocoai/nanoclaw

Path

.claude/skills/add-onecli/payload/container/skills/onecli-gateway/SKILL.md

Reviewed

Table of Contents

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.