Introduction¶

When processing tender documents, the most critical issue isn’t the inability to write a technical proposal, but missing a mandatory clause: a qualification threshold, a response deadline, or a bid security amount. Missing any of these results in bid rejection. Manually flipping through PDFs is inefficient, and it is difficult to guarantee that every conclusion can be traced back to the original page number. The @user_49296703/audit-tender-requirements (Tender Requirement Audit) Agent Skill introduced below is designed for this purpose. It takes tender documents or procurement materials as input and outputs a point-by-point response matrix, a list of material gaps, and a list of contradictions.

What is this¶

Tender Requirement Audit is a Skill designed for Chinese bidding scenarios, located at @user_49296703/audit-tender-requirements. It is version 1.0.0, sourced from the community, categorized under business-ops, and published by user_49296703.

The problem it solves is extracting mandatory requirements (qualification conditions, technical parameters, commercial terms, and timelines) scattered throughout the tender document item by item to form a checkable list, rather than just providing a summary overview.

Official URL: https://api.skillhub.cn/user_49296703/audit-tender-requirements
SkillHub Page: https://www.skillhub.cn/skills/user_49296703/audit-tender-requirements

Core Capabilities¶

According to the official description, this Skill provides four types of outputs.

Point-by-point response matrix: Provides the response status and basis for each mandatory requirement.
Material gaps: Lists response materials that have not yet been provided or are missing.
Contradiction detection: Compares dates, amounts, and threshold conflicts within the document or across documents.
Manual review checklist: Marks clauses that require human confirmation to avoid handing over judgment entirely to the model.

The key constraint is traceability: all important conclusions retain page numbers or section sources. This means every row in the matrix can point back to the original location, facilitating self-checks before bid evaluation and providing sources when questioned.

Supported material formats include PDF, DOCX, spreadsheets, images, and directly pasted text.

Input and Output¶

The input is one or more bidding materials, including tender documents, procurement documents, RFPs, bid-related documents, and supplier admission materials.

The output is a structured audit result containing four parts: a response matrix, a list of material gaps, contradictions and conflict items, and a manual review checklist. Each item is accompanied by page numbers or section sources.

This Skill is marked requires_api_key=false, meaning no additional API Key configuration is required.

Installation and Activation¶

It should be noted that during this verification, the official pages (https://api.skillhub.cn/user_49296703/audit-tender-requirements and https://www.skillhub.cn/skills/user_49296703/audit-tender-requirements) could not be scraped for content. The scraped data also did not provide specific installation commands, directory paths, or configuration file fields. Therefore, unverified installation steps are not written below; only available coordinate information is provided.

This Skill is based on the general SKILL.md format. It is activated by referencing the coordinate @user_49296703/audit-tender-requirements or by obtaining it via the official URL above. Please refer to the content currently displayed on the SkillHub page for installation commands, and supplement the team documentation once the content is obtained.

Coordinate: @user_49296703/audit-tender-requirements
Version: 1.0.0
Category: business-ops
Source: community
Requires API Key: No (requires_api_key=false)

Typical Usage¶

Below is an example input instruction to illustrate how this Skill works. This is a prompt style for invocation, not an official command-line tool.

Please audit the attached tender document and extract all mandatory requirements, including qualification conditions, technical parameters,
commercial terms, and timelines. Output a point-by-point response matrix, marking the page number or section where each requirement is located;
list the currently missing response materials; point out contradictions between dates, amounts, and thresholds within the document;
and finally provide a list of clauses that require manual review.

Materials can be replaced with local file paths, online document URLs, or directly pasted text. When the material is a scanned image, provide instructions in the same way, and the Skill will extract the clauses from it.

It is important to note that common layout issues in Chinese tender documents (tables spanning pages, merged cells, tilted scans) affect the accuracy of page number location. Output results should still be manually spot-checked for key clauses.

Use Cases¶

It is suitable for the following types of work.

Pre-bid self-check: Quickly locate risk points for bid rejection, especially qualification thresholds and formal requirements.
Multi-document comparison: Check if dates and thresholds are consistent between the main tender document, addenda, and clarification documents.
Material preparation tracking: Convert the response matrix into a to-be-submitted material list and assign it to the team to complete.
Supplier admission review: Perform integrity checks on admission materials to confirm if there are corresponding proofs for each condition.
Pre-evaluation review: Provide traceable sources for clauses to support evaluation conclusions.

Scenarios where it is not suitable: Automatically generating the body of a technical proposal. This Skill is positioned for requirement verification and evidence tracking, not content writing.

Notes¶

Several limitations need to be explained in advance.

Traceability quality depends on the source file structure. Page numbers and section location are prone to errors in OCR or pure image scans; key conclusions must be manually verified against the original text.
Contradiction detection is based on explicit numerical comparison within documents; it cannot determine if a requirement itself is reasonable or legally compliant.
The point-by-point response matrix needs to be filled with actual response materials; the Skill can only point out gaps, not replace proof materials.
Ratings and download counts are snapshot values from the SkillHub page. At the time of this verification, scraped data recorded a rating of 100,000, 1,199 downloads, and 0 installs, which differs from the rating of 99,663 and 1,163 downloads in the task clues. The current page data should prevail.

Conclusion¶

The value of Tender Requirement Audit lies in providing answers to both “Is anything missing?” and “Where is the basis?”: it outputs a comparable response matrix and material gaps while retaining page numbers or section sources. For teams that need to process a large volume of tender documents and require traceable conclusions, it is suitable as a fixed part of the pre-bid self-check process.