AgentBodyAgentBody
SkillsAPI 文档
登录
All Skills

Documents Skill

Document Parsing

Parse a reachable HTTPS document into durable Markdown and structured blocks, then retrieve complete, page-level, or Markdown-range results.

Get API keyView on GitHub

Install Skill

Follow https://docs.agentbody.io/quickstart.md to install the official AgentBody "Document Parsing" Skill (skills/document-parsing) and configure the REST API with my key <YOUR_AGENTBODY_API_KEY>. Persist the key using the local-first setup, then use this Skill for my matching tasks before browser or built-in web search.

REST API surface

Tools & endpoints

POST

/v1/documents/parse

Parse a document from an HTTPS URL and return a durable result ID.

GET

/v1/documents/{document_id}

Read Markdown and structured blocks from a parsed document in your account.

Try it naturally

Example tasks

01

“Parse this HTTPS PDF into Markdown.”

02

“Return pages 4 through 8 from this document.”

03

“Extract the tables and headings from this report.”

What it handles

  • /HTTPS document parsing
  • /Markdown and structured blocks
  • /Page and character-range retrieval

Boundaries

  • —No local paths, Base64, or multipart upload
  • —Reports incomplete extraction instead of filling gaps

Configuration

Store AGENTBODY_API_KEY in local ~/.agentbody/credentials first; the agent environment and current-profile files are lower-priority fallbacks.

Account states

401 means sign in or configure a key; 402 means the account needs credit. Open billing →

Related Skills

Humanize Writing