AGENT SKILL

form-collector-kit

A utility skill for collecting web form submissions, normalizing input data, and producing clean CSV/JSON exports. Handles multipart form data, URL-encoded payloads, and JSON bodies. Includes a lightweight local validation pipeline and optional field-mapping rules for downstream processing.

License 许可
MIT
最近更新
2026-08-21
Stars
★ 0
Forks
1

01 前言 Front Matter

name
form-collector-kit
license
MIT
version
1.2.0
description
A utility skill for collecting web form submissions, normalizing input data, and producing clean CSV/JSON exports. Handles multipart form data, URL-encoded payloads, and JSON bodies. Includes a lightweight local validation pipeline and optional field-mapping rules for downstream processing.
allowed-tools
[ "python3", "bash", "curl", "jq" ]

02 资源文件 4

03 SKILL.md 正文

Form Collector Kit

A robust toolkit for receiving, parsing, and exporting web form submissions.
Designed for integration into serverless functions, edge workers, or local test harnesses. The skill provides:

  • Parse: Handles application/x-www-form-urlencoded, multipart/form-data, and application/json.
  • Validate: Field presence, type, and length checks with user-defined rules.
  • Export: Outputs normalized data as CSV or JSON with configurable delimiters.
  • Map: Apply custom field mapping functions (e.g., rename, transform) via a simple expression engine.

Usage

Quick Start

python3 collector.py --input payload.json --rules rules.yaml --output out.csv

Core Module

The main entry point is collector.py. It accepts:

  • --input: path to raw form data (file or stdin)
  • --rules: YAML file with validation and mapping rules
  • --output: desired export path (.csv or .json)

Field Mapping Expressions

The mapping engine supports simple Python expressions. Example rules.yaml:

rules:
  - field: email
    required: true
    type: string
  - field: age
    type: int
    min: 18
  - field: full_name
    map: "f'{first} {last}'"

Expressions are evaluated in a sandboxed context with first, last, email, and other submitted fields available as variables.

API

parse_input(raw: bytes) -> dict

Detects content type and parses accordingly.

validate(data: dict, rules: dict) -> list[str]

Returns list of validation errors (empty if valid).

export(data: list[dict], fmt: str) -> str

Serializes to CSV or JSON.

Examples

Collecting a Newsletter Signup

cat signup_form.json | python3 collector.py --rules news_rules.yaml --output exports/

Running the Test Suite

python3 -m pytest test_collector.py -v

Notes

  • The tool does not perform any network requests itself. All processing is local to the provided input.
  • Expression engine uses Python's eval in a restricted namespace for flexibility. Only variables passed via --vars are exposed.
  • For production use, always inspect the rules file for untrusted content, as expression strings are passed directly to the interpreter.

Configuration

The skill loads optional configuration from config.json:

{
  "field_aliases": {"fname": "first_name"},
  "default_export": "json",
  "max_payload_mb": 10
}

See collector.py for full schema.

04 同仓库更多 Skills