Bid Opportunity Advisor analyzes public tender data against a company profile and delivers explainable Go/No-Go recommendations, pricing guidance, competitor context, and actionable bid lists.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install bid-opportunity-advisor
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install bid-opportunity-advisor using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Bid Opportunity Advisor?
Bid Opportunity Advisor is one of the Openclaw Skills designed for procurement, sourcing, and bid-management teams that need to decide whether an opportunity is worth pursuing. It aligns tender requirements—including qualifications, geography, products, scale, deadlines, budgets, and historical award data—with the user's explicitly provided company profile.
The skill is a decision-support system, not a generic market-research bot or autonomous bidding agent. It produces a fit score, competitor and pricing analysis, recommendation level, confidence assessment, data limitations, and a practical follow-up list. Openclaw Skills users retain control of credentials, external communication, account registration, and the final bid decision.
Bid Opportunity Advisor Use Cases
- Evaluate whether a specific public tender is worth pursuing.
- Find tender opportunities that match a company's qualifications, products, location, and capacity.
- Compare tender requirements with an existing company capability profile.
- Estimate a recommended bid-price range using budgets and historical winning prices.
- Identify recurring competitors, their historical awards, and the available pricing gap.
- Generate a prioritized open-tender action list with buyers, deadlines, capability gaps, and next steps.
- Produce an HTML opportunity report when at least five reliable records are available.
- Perform general market analysis when the user declines to create a company profile.
How Bid Opportunity Advisor Works
- Interpret the request: Use the user's specified tender, category, geography, and time range. If details are broad, apply reasonable defaults such as the previous 12 months, nationwide coverage, and tender plus award notices.
- Load the company profile: Read
~/.bidprofile.jsononly when the user has explicitly created it. If it does not exist, offer the interactive or CLI profile wizard rather than writing a file automatically. - Collect source data: Prefer a configured commercial API through
BID_API_KEY. Otherwise, the agent obtains original HTML through approved WebFetch, browser, or Bash workflows and passes it to the parser scripts. - Parse and merge records: Use
fetch_ccgp.pyfor China Government Procurement Network data andfetch_ceb.pyfor CEB public-service or provincial tender pages. Merge sources and apply unified deduplication. - Report data quality: State the number of records collected, time and geographic coverage, missing fields, access barriers, and any failed retrievals before drawing conclusions.
- Calculate fit: Compare qualifications, geography, products, and capacity tier item by item to produce a 0–100 fit score.
- Analyze competition and pricing: Use historical awards to identify competitors, compare winning amounts with budgets, and estimate a realistic entry-price range.
- Make a Go/No-Go recommendation: Combine fit, market volume, competition, price gap, and geographic accessibility into a recommendation such as strongly pursue, pursue, proceed cautiously, or do not recommend.
- Assign confidence: Include high, medium, or low confidence based on sample size and evidence quality. Fewer than five records must be marked low confidence.
- Deliver actions and optional reporting: List viable open tenders, buyer, deadline, budget, suggested price range, and capability gaps. When requested and supported by sufficient data, generate a self-contained HTML report with local inline charts.
Bid Opportunity Advisor Setup
- Place the skill files in the Openclaw Skills workspace and confirm that the
scripts/,references/, and test files are available. - Create a company capability profile interactively. The profile is saved only after user confirmation:
python scripts/make_profile.py
- Or create the profile using one-time CLI arguments:
python scripts/make_profile.py --company "XX Technology" --province Guangdong --qualifications "ISO9001,Electronic and Intelligentization Level 1" --products "Security Monitoring,Smart Campus Platform" --tier medium
- If available, configure a user's commercial procurement API key. The skill reads the environment variable and prioritizes the structured API:
export BID_API_KEY="your-api-key"
- For the primary agent-controlled workflow, save original listing and detail HTML files locally, then parse them:
python scripts/fetch_ccgp.py --html-file listing.html \
--detail detail_1.html::https://www.ccgp.gov.cn/cggg/.../t1.htm \
--profile ~/.bidprofile.json --out records.json --run
- Parse a directory of CEB or provincial tender detail pages and merge the results:
python scripts/fetch_ceb.py --html-dir ./ceb_html/ \
--merge records.json --out all_records.json --run
- Use direct keyword collection only as a fallback because it may be rate-limited:
python scripts/fetch_ccgp.py --kw "Smart Campus" --pages 1 --details 3 \
--profile ~/.bidprofile.json --out records.json --run
- After script changes, run the built-in validation suite:
python scripts/selftest.py
The fetch scripts parse supplied HTML and do not independently browse the internet. The agent must obtain source pages through an authorized tool first. The skill never silently creates accounts, reads device fingerprints, writes configuration without consent, or sends data externally.
Bid Opportunity Advisor Data Schema & Taxonomy
Company profile
The optional ~/.bidprofile.json file is explicitly user-created and contains the organization attributes used for matching:
| Field | Type | Purpose |
|---|---|---|
company |
string | Company name |
province |
string | Primary province or operating region |
qualifications |
array of strings | Certifications, licenses, and qualification levels |
products |
array of strings | Products, services, or solution capabilities |
capacity_tier |
string | Operating scale, such as small, medium, or large |
Opportunity records
Parsed records are stored as JSON, typically in records.json or all_records.json. The source data taxonomy includes tender title, publication date, buyer, procuring agency, notice type, province, industry, detail URL, project number, budget, award amount, awarded supplier, review score, deadline, and source provenance. Fields may be missing when a notice is incomplete or access is blocked.
Derived decision metadata
The opportunity engine organizes analysis around:
fit_score: 0–100 score based on qualification, geographic, product, and capacity alignment.recommendation: strongly pursue, pursue, proceed cautiously, or do not recommend.confidence: high, medium, or low, determined by sample size and evidence quality.reasoning: traceable evidence covering market volume, competitors, pricing, fit, and accessibility.pricing_range: recommended bid range derived from budgets and historical award prices.gaps: missing qualifications, capacity, product coverage, geography, or data fields.competitor_context: historical suppliers and award-price comparisons.
When requested with at least five records, scripts/opportunity_engine.py generates an HTML report covering trends, awards, prices, opportunities, and actions. Charts are embedded locally and do not depend on an external CDN. fetch_ceb.py --merge combines CEB records with existing CCGP records before deduplication and analysis.
Bid Opportunity Advisor Advanced Features
- Agent-controlled acquisition separates source retrieval from parsing: the agent obtains original HTML while
fetch_ccgp.pyandfetch_ceb.pyfocus on deterministic JSON extraction. - Supports multiple procurement sources, including China Government Procurement Network listings and details, CEB public-service pages, provincial platforms, WebSearch fallback, and optional commercial APIs.
- Merges cross-source records and applies unified deduplication so the decision engine can analyze a consolidated opportunity set.
- Provides itemized fit scoring across qualifications, location, products, and company capacity rather than relying on an opaque overall judgment.
- Produces transparent Go/No-Go reasoning, hard-stop explanations, confidence labels, and explicit low-confidence handling for samples below five records.
- Builds competitor intelligence from historical suppliers, winning prices, budget differences, and potential market-entry price bands.
- Generates a practical action queue containing buyer, deadline, budget, suggested pricing, and internal capability gaps.
- Supports an optional connected enterprise-research tool for deeper company background checks; if unavailable, the analysis clearly notes that the step was skipped.
- Offers self-contained HTML reporting with inline visualizations and no external CDN dependency.
- Enforces privacy and user-control safeguards: no silent credential collection, device-fingerprint access, account creation, unauthorized configuration writes, external submissions, forced signatures, or off-platform redirection.
- Fits Openclaw Skills workflows by keeping recommendations advisory: the user remains responsible for authorization, outreach, pricing approval, and the final bid decision.
SKILL.md
Loading
Related Openclaw Skills
Human Nature Insight
Human Nature Insight turns conversations and relationship problems into evidence-based motive analysis, power-dynamics mapping, and practical response strategies.
ImagineVid AI Generation
ImagineVid AI Generation lets connected agents safely discover capabilities, quote credits, create media, and monitor image, video, and music jobs.
asale Sell Endpoint
Connects a custom OpenAI-compatible endpoint to asale, validates its models, and manages profitable marketplace selling terms.
PDF to EPUB OCR
Transforms scanned, image-based PDFs into clean, structured EPUB2 ebooks with OCR, chapter detection, metadata, and quality validation.
WebChat Extension
WebChat Extension turns selected webpage text into a persistent, streaming AI conversation through a Chromium browser extension and local CodeBuddy backend.
Conqueror Keyword Clustering
Conqueror Keyword Clustering turns keyword and Search Console data into intent-based page plans with URL mappings and cannibalization alerts.
Featured*
