← Back to Skills Marketplace
Bid Reader
by
ezhencacao-dotcom
· GitHub ↗
· v1.0.0
· MIT-0
67
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install bid-reader
Description
Extracts and returns plain text from PDF, Word (.docx), and Excel (.xlsx/.xls) bid documents for analysis, search, or summarisation.
README (SKILL.md)
bid-reader Skill
Overview
A lightweight skill to extract readable text from bid and tender documents in PDF, Word (.docx), and Excel (.xlsx/.xls) formats. It can be invoked from the OpenClaw UI or other agents to quickly pull the full textual content of a file for analysis, search, or summarisation.
Usage
bid-read \x3Cfile-path>
\x3Cfile-path>should be an absolute or workspace‑relative path to a document.- The skill prints the extracted plain‑text to stdout, which OpenClaw captures and returns to the caller.
Example
bid-read /home/zhenxing/投标文件/招投标项目1/13.上海联通/投标文件.pdf
The command returns the full text of the PDF, ready for further processing (e.g., keyword search, summarisation).
Installation
Copy the skill folder into your workspace under skills/bid-reader. Install required Python packages:
pip install -r $(pwd)/skills/bid-reader/requirements.txt
The skill is then available as an agent command.
Implementation Details
- PDF: Uses
pdfplumberto extract text page‑by‑page. - Word: Uses
python-docxto read paragraphs. - Excel: Uses
pandas(withopenpyxl/xlrd) to read all sheets and concatenate cell values.
Limitations
- Only
.pdf,.docx,.xlsx, and.xlsare supported. Other formats will be ignored. - Large files may take a few seconds to process.
- Tables are flattened into whitespace‑separated rows; complex formatting is not preserved.
Future Enhancements
- Add OCR fallback for scanned PDFs (e.g., via
pytesseract). - Support selective page or sheet extraction.
- Provide a JSON output mode with structural metadata.
Usage Guidance
This skill appears internally consistent and limited to extracting text from files. Before installing: (1) install the Python deps in an isolated virtual environment to avoid polluting your system/site packages, (2) be aware the script will read any file path you pass it — do not point it at sensitive files you don't want processed, (3) the repo does not provide a 'bid-read' executable wrapper; run the script directly (python scripts/read.py /path/to/file) or add your own launcher, and (4) note that .doc support is advertised but likely unsupported by python-docx (use .docx). If you need OCR for scanned PDFs or stricter sandboxing, add those safeguards before using on untrusted documents.
Capability Assessment
Purpose & Capability
Name/description match the code and requirements: pdfplumber, python-docx, pandas/openpyxl are appropriate for extracting text from PDF, .docx and Excel files. Minor mismatch: SKILL.md advertises a 'bid-read' command but no wrapper is provided in the bundle (you'd need to invoke scripts/read.py or create an executable entrypoint). The code also accepts '.doc' extension but uses python-docx (which only supports .docx), so '.doc' handling is likely to fail — a correctness issue, not a security one.
Instruction Scope
SKILL.md instructs the agent to read a workspace or absolute file path and print extracted text to stdout. The implementation only reads the provided file and does not access other files, network endpoints, or environment variables. There is no broad or vague instruction to gather extra context or scan unrelated paths.
Install Mechanism
No packaged install spec is provided (instruction-only plus a local script). Installation relies on pip installing requirements.txt from PyPI, which is normal for a Python utility but will fetch third-party packages over the network; consider installing in a virtual environment. Nothing in the install steps pulls code from arbitrary URLs or personal servers.
Credentials
The skill declares no required environment variables, credentials, or config paths and the code does not read any env vars. Requested permissions are minimal and proportional to the task.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request permanent presence or modify other skills or system-wide settings. Autonomous invocation is allowed by default but the skill's operations are limited to local file reading when invoked.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install bid-reader - After installation, invoke the skill by name or use
/bid-reader - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of bid-reader skill.
- Extracts readable text from PDF, Word (.docx), and Excel (.xlsx/.xls) bid/tender documents.
- Simple command-line usage: bid-read <file-path>.
- Outputs plain text for search, summarisation, or further analysis.
- Supports invocation from OpenClaw UI or other agents.
- Handles PDF via pdfplumber, Word via python-docx, Excel via pandas/openpyxl/xlrd.
- Ignores unsupported formats; flattens tables and does not preserve complex formatting.
Metadata
Frequently Asked Questions
What is Bid Reader?
Extracts and returns plain text from PDF, Word (.docx), and Excel (.xlsx/.xls) bid documents for analysis, search, or summarisation. It is an AI Agent Skill for Claude Code / OpenClaw, with 67 downloads so far.
How do I install Bid Reader?
Run "/install bid-reader" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Bid Reader free?
Yes, Bid Reader is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Bid Reader support?
Bid Reader is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Bid Reader?
It is built and maintained by ezhencacao-dotcom (@ezhencacao-dotcom); the current version is v1.0.0.
More Skills