CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

vercel-connect

Vercel Connect expert guidance for securely obtaining scoped credentials for third-party services on behalf of apps or users. Use when wiring up provider API access, OAuth, API-key services, MCP servers, triggers, framework adapters, eve agent connections, or provider tokens for Passport-authenticated users.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

77%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 with executable code for CLI, SDK, HTTP, and every adapter path, and the recommended workflow is clearly sequenced with genuine error-recovery feedback loops. Its weaknesses are efficiency and structure: the eve channel sections repeat the same explanations verbatim, version pins are scattered without a deprecation section, and everything lives inline in SKILL.md with no progressive disclosure to reference files.

Suggestions

Extract the per-channel helper details (connectSlackCredentials, connectGitHubCredentials, connectLinearCredentials) and the framework adapter configs (Better Auth, Auth.js) into reference files (e.g. references/eve-channels.md, references/adapters.md), keeping only a one-line summary and a clearly signaled link per item in SKILL.md.

Deduplicate the thrice-repeated webhookVerifier/botToken explanation by stating the shared 'Connect forwards verified provider webhooks as signed Vercel OIDC requests verified by vercelOidc()' pattern once and referencing it from each channel section.

Consolidate the scattered version pins ('CLI 58.1+', 'CLI 64.1.0+', '@vercel/connect 2.4.0+') into a single 'Version requirements' or 'Old patterns' section so time-sensitive details don't penalize the main flow.

DimensionReasoningScore

Conciseness

The body is mostly efficient, product-specific knowledge Claude cannot be assumed to know, but the three eve channel sections repeat the near-identical 'webhookVerifier: a Vercel OIDC verifier (vercelOidc()) ... instead of the raw ... secret' explanation verbatim three times, and scattered version pins ('CLI 58.1+', 'CLI 64.1.0+', '@vercel/connect 2.4.0+') are time-sensitive without an 'old patterns'/'deprecated' section. This fits anchor 3's 'mostly efficient but could be tightened', below anchor 4's 'minor instances' and well above anchor 2.

3 / 5

Actionability

Guidance is fully executable and copy-paste ready: a complete getToken + fetch Slack example, the full CLI lifecycle (list/create/attach/token), a working Python HTTP example, eve connection and channel definitions, and betterauth/authjs provider configs. This matches anchor 5's 'copy-paste ready code or commands; specific examples cover the common cases'.

5 / 5

Workflow Clarity

The 'Recommended workflow' gives an explicit 8-step sequence with validation steps (capture the returned UID, verify trigger delivery via event history and correlation IDs, wait for authorization completion) and feedback loops for error recovery (re-run 'vercel env pull' when dev OIDC tokens expire; call startAuthorization() after UserAuthorizationRequiredError). This matches anchor 5; no destructive or batch operations trigger the validation cap.

5 / 5

Progressive Disclosure

The body is well-sectioned with clear headers, but it is a ~317-line monolith with no internal reference files: per-channel helper details (Slack/GitHub/Linear) and per-adapter configuration that would fit separate reference files are fully inlined, and only external doc URLs in 'Sources of truth' serve as references. This matches anchor 3's 'content that should be separate is inline' rather than anchor 4, where most content is appropriately placed across files.

3 / 5

Total

16

/

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 states a concrete capability, includes an explicit and detailed 'Use when...' trigger clause, and is tightly scoped to the Vercel Connect product. Keyword coverage and action enumeration are good but not exhaustive, and a few generic terms leave minor overlap risk with general OAuth/MCP skills.

DimensionReasoningScore

Specificity

Names the domain ('securely obtaining scoped credentials for third-party services on behalf of apps or users') and enumerates several concrete scenarios (OAuth, API-key services, MCP servers, triggers, framework adapters, eve agent connections, Passport provider tokens), matching anchor 4's 'several specific actions; minor gaps'. It falls short of anchor 5 because the coverage is scenario framing rather than a comprehensive multi-action inventory, and is clearly above anchor 3's 1-2 concrete actions.

4 / 5

Completeness

Explicitly answers 'what' ('securely obtaining scoped credentials for third-party services on behalf of apps or users') and 'when' via a literal 'Use when wiring up provider API access, OAuth, ... MCP servers, triggers, framework adapters, eve agent connections, or provider tokens for Passport-authenticated users' clause with concrete trigger phrases — a direct match for anchor 5.

5 / 5

Trigger Term Quality

Includes natural terms users would say — 'provider API access', 'OAuth', 'API-key services', 'MCP servers', 'triggers', 'eve agent connections' — giving good keyword coverage per anchor 4. Not anchor 5 because common user phrasings like 'slack token', 'connect to Slack', or 'get a GitHub token' are absent from the description itself (they only appear in frontmatter metadata).

4 / 5

Distinctiveness Conflict Risk

Occupies a clear product-specific niche (Vercel Connect, eve, Passport) with distinct triggers, but broad terms like 'OAuth' and 'MCP servers' create minor overlap risk with generic auth/MCP skills — anchor 4's 'mostly distinct; minor overlap risk' rather than anchor 5's minimal conflict risk.

4 / 5

Total

17

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository

vercel/vercel-plugin

Path

skills/vercel-connect/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.