Structured Extraction

What it does

When your agent uses it

• Tasks that involve Document parsing

Example prompts

• “/structured-extraction”

Requirements

• Python 3

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1Read the ## Current attachments section in the system prompt. It gives each original filename plus its preprocessed Markdown and…
2. 2For each source, call document_conversion_api_extractdocument directly with run_python only when the preprocessed Markdown is missing or a…
3. 3In that Python call, serialize the complete result to /home/user/work/.json instead of printing it. For example
4. 4Prefer reading or searching the preprocessed /home/user/attachments//content.md with read_file or run_bash commands such as grep; it…
5. 5Preserve the source filename and every available location marker when reporting evidence. Keep extraction separate from interpretation.
6. 6If conversion fails or a format is unsupported, report the exact failure and do not substitute guessed content.

What it can do on your machine

Read from SKILL.md and the folder at commit c155c00. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Structured Extraction loads about 579 tokens when it runs. Until then it costs about 35 tokens; SKILL.md has 241 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~35

When it runs · the whole SKILL.md, loaded when a task matches

~579

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from bionic-gpt/bionic-gpt at commit c155c00, republished under its Apache-2.0 licence (© bionic-gpt). 241 words, ~579 tokens.

name

structured-extraction

description

Extract structured, source-located evidence from uploaded documents with Xberg for document analysis and validation.

import json

result = document_conversion_api_extractdocument(
    file_path="/home/user/attachments/<attachment-id>/original/<source-file>"
)
with open("/home/user/work/<source-file>.json", "w") as extracted:
    extracted.write(json.dumps(result, ensure_ascii=False, indent=2))
print("Saved extraction to /home/user/work/<source-file>.json")

© bionic-gpt, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in crates/builtin-skills/skills/structured-extraction of bionic-gpt/bionic-gpt.

Open the folder on GitHubat commit c155c00