Skill Allowed Tools Array
skill-allowed-tools-array
info
experience: securityexperience: governance
spec: frontmatter
allowed-tools must be an array of tool names.
If a skill declares allowed-tools, it must be an array of tool names so its capability surface is explicit and auditable.
Rule definition
The full Spotlight rule — drop it into a ruleset (.spotlight.yaml / .spectral.yaml) and run it with any Spotlight tool.
skill-allowed-tools-array:
title: Skill Allowed Tools Array
reference: https://spotlight-rules.com/spec/rules/agent-skill/skill-allowed-tools-array/
description: If a skill declares allowed-tools, it must be an array of tool
names so its capability surface is explicit and auditable.
message: allowed-tools must be an array of tool names.
severity: info
given: $.frontmatter['allowed-tools']
then:
function: schema
functionOptions:
schema:
type: array
items:
type: string
tags:
- format:agent-skill
- spec:frontmatter
- experience:security
- experience:governance
prompt: "You are editing an Agent Skill document to satisfy the Spotlight API
governance rule 'skill-allowed-tools-array' (Skill Allowed Tools Array).
Requirement: If a skill declares allowed-tools, it must be an array of tool
names so its capability surface is explicit and auditable. To fix: Adjust
the targeted value so it conforms to the schema this rule requires. This
rule is evaluated at the JSONPath `$.frontmatter['allowed-tools']` — inspect
every location it matches and correct only what violates the rule. Make the
smallest change that satisfies the rule, leave all unrelated content, key
order, comments, and formatting unchanged, and keep the document valid Agent
Skill. Return only the complete corrected document, with no commentary."
AI fix prompt — send this to Claude, Gemini, or ChatGPT with your artifact to auto-fix this rule
You are editing an Agent Skill document to satisfy the Spotlight API governance rule 'skill-allowed-tools-array' (Skill Allowed Tools Array). Requirement: If a skill declares allowed-tools, it must be an array of tool names so its capability surface is explicit and auditable. To fix: Adjust the targeted value so it conforms to the schema this rule requires. This rule is evaluated at the JSONPath `$.frontmatter['allowed-tools']` — inspect every location it matches and correct only what violates the rule. Make the smallest change that satisfies the rule, leave all unrelated content, key order, comments, and formatting unchanged, and keep the document valid Agent Skill. Return only the complete corrected document, with no commentary.
| Artifact | Agent Skill |
|---|---|
| Applies at (given) | $.frontmatter['allowed-tools'] |
| Severity | info — educational by default; raise to warn/error to enforce |
| Reference | https://spotlight-rules.com/spec/rules/agent-skill/skill-allowed-tools-array/ |