veryfront/skill

Import

import {
  createExecuteSkillScriptTool,
  createLoadSkillReferenceTool,
  createLoadSkillTool,
  filterToolNamesForSkill,
  filterToolsForSkill,
  getAllSkills,
} from "veryfront/skill";

Examples

import { parseSkillFrontmatter, validateSkillMetadata } from "veryfront/skill";

const parsed = await parseSkillFrontmatter("---\nname: review\ndescription: Review code\n---\n");
validateSkillMetadata(parsed.frontmatter, "review");

Exports

Components

Name	Description	Source
`SKILL_ALLOWED_TOOL_PATTERN_REGEX`	Public inspection matcher for exact tool IDs and prefix wildcards. Mutating this compatibility value does not alter authorization decisions.	source
`SKILL_ASSETS_DIR`		source
`SKILL_DESCRIPTION_MAX_LENGTH`	Maximum description length in characters	source
`SKILL_MD_FILENAME`	Standard SKILL.md filename per agentskills.io spec	source
`SKILL_METADATA_KEY_MAX_LENGTH`		source
`SKILL_METADATA_MAX_ENTRIES`		source
`SKILL_METADATA_VALUE_MAX_LENGTH`		source
`SKILL_NAME_REGEX`	Historical public skill-name inspection matcher. Mutating this compatibility value does not alter framework admission.	source
`SKILL_READABLE_DIRS`	Canonical read-only skill directories exposed through reference loading.	source
`SKILL_REFERENCES_DIR`		source
`SKILL_RESOURCES_DIR`		source
`SKILL_SCRIPTS_DIR`	Conventional subdirectory names	source
`SKILL_TOOL_IDS`	Public snapshot of tool IDs that belong to the skill system.	source

Functions

Name	Description	Source
`createExecuteSkillScriptTool`	Create the execute_skill_script tool. Executes a script from a skill’s scripts/ directory.	source
`createLoadSkillReferenceTool`	Create the load_skill_reference tool. Reads a reference file from a skill’s references/, resources/, or assets/ directory.	source
`createLoadSkillTool`	Create the load_skill tool. Loads a skill’s full instructions, available references, and scripts.	source
`filterToolNamesForSkill`	Filter provider-native or other name-only tool inventories through the same boundary.	source
`filterToolsForSkill`	Filter tool definitions before sending them to the model.	source
`getAllSkills`		source
`getSkill`		source
`getSkillScriptExecutor`	Get the appropriate script executor. Checks cloud auth availability on every call so request-scoped credentials and environment overrides are respected.	source
`isSkillInfrastructureToolId`	Framework-owned membership check that cannot be changed by public Set mutation.	source
`isSkillToolAvailable`	Check whether a specific tool call is available at execution time.	source
`isSkillVisibleTo`	Whether a skill is visible to the caller identified by the scope.	source
`isValidProviderSafeSkillId`	Framework-owned provider-safe owned skill-id grammar check.	source
`isValidSkillName`	Framework-owned historical skill-name grammar check.	source
`isValidStrictSkillName`	Framework-owned strict filesystem skill-name grammar check.	source
`listSkillSubdir`	List files with the public compatibility resource policy. Enumeration is not entry-capped and preserves the filesystem adapter’s iteration order. `listStrictSkillSubdir` applies the runtime filesystem ceilings and deterministic ordering.	source
`parseBoundedSkillDocument`	Parse one bounded Skill document with an explicit provider or the active extension contract generation.	source
`parseSkillFileFrontmatter`	Parse and bound an untrusted SKILL.md document read from a filesystem boundary. YAML frontmatter is decoded by the explicit provider, or by the active `SkillDocumentParserProvider` registration when the argument is omitted.	source
`parseSkillFrontmatter`	Parse SKILL.md content through the bounded, fail-closed format. Malformed YAML, invalid Unicode, and oversized documents are rejected. YAML frontmatter is decoded by the explicit provider, or by the active `SkillDocumentParserProvider` registration when the argument is omitted.	source
`registerSkill`		source
`validateSkillFileMetadata`	Validate metadata loaded from a filesystem skill. The caller-supplied directory/runtime identity remains canonical; a differing authored `name` is display metadata and never participates in lookup or authorization.	source
`validateSkillMetadata`	Validate and normalize parsed frontmatter into SkillMetadata.	source
`validateSkillPath`	Validate a requested path with the public compatibility resource policy. Relative paths may contain up to 4096 characters and filesystem directory enumeration is not entry-capped. `validateStrictSkillPath` applies the runtime filesystem ceilings.	source

Types

Name	Description	Source
`ActiveSkillContext`	Active skill context for runtime availability and delegation tracking	source
`AgentCapabilityScope`	Caller scope used for owner-aware capability resolution.	source
`ParsedSkillContent`	Result of splitting and decoding one bounded `SKILL.md` document.	source
`Skill`	Registered skill instance	source
`SkillContent`	Full skill content returned by load_skill tool	source
`SkillMetadata`	Parsed frontmatter metadata from SKILL.md	source
`SkillScriptExecutor`	Script executor interface.	source
`SkillScriptExecutorInput`	Input for the script executor	source
`SkillScriptResult`	Result from executing a skill script.	source
`SkillScriptSnapshot`	Bounded, validated script tree used to preserve same-directory imports.	source
`SkillScriptSnapshotFile`	One validated text file retained in an executable skill-script snapshot.	source

Constants

Name	Description	Source
`skillRegistry`		source

Getting Started

Concepts

Guides

API Reference

Import

Examples

Exports

Components

Functions

Types

Constants

​Import

​Examples

​Exports

​Components

​Functions

​Types

​Constants

Import

Examples

Exports

Components

Functions

Types

Constants