CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

vercel-agent

Vercel Agent guidance — dashboard and Slack chat, code review, Security Review, production investigation, approved actions, and product installation. Use when configuring or working with Vercel's AI assistant.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/vercel-agent/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

65%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 compact, well-organized product-knowledge skill that respects token budget and navigates cleanly, with clear external references. Its weaknesses are instructional: most content describes what Vercel Agent does rather than telling Claude what to do, and task workflows (e.g., triaging an investigation or reading security findings end-to-end) are left implicit.

Suggestions

Add concrete example invocations for the commands you name — e.g., a sample 'vercel security vulnerabilities' call with expected output — so guidance is executable rather than descriptive.

Replace some capability descriptions with scenario-to-action mappings (e.g., under When to Use: 'anomaly alert fired → start an Investigation from the dashboard Agent chat'), giving Claude an explicit workflow for the common multi-step cases.

Deduplicate the pricing facts: state the 10-investigations limit and Installation's zero charge once each, in the Pricing section, instead of repeating them in the Investigation and Installation sections.

DimensionReasoningScore

Conciseness

Mostly lean, product-specific facts with no explanation of concepts Claude already knows, but two facts are stated twice — 'Observability Plus includes 10 investigations per billing cycle' appears in both Investigation and Pricing, and 'Free (no credit cost)' duplicates 'Installation has no Agent charge' — matching anchor 4's minor trimming opportunity.

4 / 5

Actionability

A few concrete anchors exist ('Select Agent in the top-right corner', '@vercel mention in PR comments', 'vercel security vulnerabilities'), but the bulk describes capabilities ('Analyzes anomaly alerts, failed deployments, runtime errors...') rather than instructing, and the single CLI command is named without an example invocation — anchor 3's 'some concrete guidance but incomplete'.

3 / 5

Workflow Clarity

The body is organized by capability topic rather than task flow; 'When to Use' maps scenarios only tersely and just one section (Security Review) connects a scenario to an actual command. No multi-step sequence with checkpoints is present, and the simple-skill exception doesn't apply to this multi-capability skill.

3 / 5

Progressive Disclosure

With no bundle files, the ~60-line body stays entirely inline, well-sectioned under clear headers, with a clearly signaled References section of one-level-deep external doc URLs and one explicit cross-skill pointer ('the vercel-cli skill's references/security.md') — nothing that belongs in a separate file is inlined.

5 / 5

Total

15

/

20

Passed

Description

75%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 solid description with a specific capability enumeration, natural trigger terms, and an explicit 'Use when' clause, all anchored to a distinct product niche. It sits just below the top anchor because its capabilities are named as feature areas rather than actions, and its when-clause and trigger coverage lack the concreteness and synonym breadth of the best examples.

DimensionReasoningScore

Specificity

Enumerates six concrete capability areas — 'dashboard and Slack chat, code review, Security Review, production investigation, approved actions, and product installation' — but as feature nouns rather than verb-phrase actions, leaving minor gaps versus the comprehensive anchor 5.

4 / 5

Completeness

Explicitly answers both 'what' (the enumerated capability list) and 'when' ('Use when configuring or working with Vercel's AI assistant'), but the when-clause is generic rather than listing concrete trigger phrases, so anchor 4 fits best.

4 / 5

Trigger Term Quality

Natural user phrases like 'code review', 'Security Review', 'production investigation', 'dashboard chat', and 'Slack' are present, but common variants such as 'PR review', 'incident', or repeated 'AI assistant' phrasing are missing, matching anchor 4 rather than 5.

4 / 5

Distinctiveness Conflict Risk

The 'Vercel's AI assistant' scoping gives it a clear niche, yet bare terms like 'code review' and the existence of closely related Vercel skills (vercel-cli, deployments-cicd) create minor overlap risk, matching anchor 4 rather than 5.

4 / 5

Total

16

/

20

Passed

Validation

75%

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

Validation — 12 / 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

referenced_paths_exist

Referenced path issues: 1 possibly missing (references/security.md)

Warning

Total

12

/

16

Passed

Repository

vercel/vercel-plugin

Path

skills/vercel-agent/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.