home / skills / pandaallin / ubos_final / eu-grant-hunter
This skill scans EU funding sources to identify grant opportunities aligned with UBOS, tracks deadlines, scores fit, and generates concise briefs.
npx playbooks add skill pandaallin/ubos_final --skill eu-grant-hunterReview the files below or copy the command above to add this skill to your agents.
---
name: eu-grant-hunter
description: |
Scans EU funding databases (Horizon Europe, ERDF, Digital Europe,
Innovation Fund) to identify grant opportunities matching UBOS
capabilities. Tracks deadlines with multi-level reminders (90/60/30/7
days), scores opportunities by fit (0-5), and generates opportunity
briefs. Use when discussing EU grants, funding opportunities, or when
tracking the €70M+ pipeline. Automatically runs daily at 09:00 UTC.
license: UBOS Constitutional License
version: 1.0.0
author: Janus-in-Claude (Architect) + Codex (Forgemaster)
created: 2025-10-30
---
# EU GRANT HUNTER
## Purpose
Scan the €70M+ EU funding pipeline for opportunities aligned with UBOS projects, calculate fit scores, and ensure no deadlines are missed. Automations keep the Trinity informed while manual queries deliver targeted intelligence on demand.
## When To Use
- Daily at 09:00 UTC for the automated grant scan
- When a query mentions grants, EU funding, Horizon Europe, ERDF, Innovation Fund, or Digital Europe
- When asked to “find grants for” a topic or project
- When a pipeline status update is requested
- When preparing opportunity briefs for Trinity review
## Core Capabilities
- Crawl Horizon Europe, ERDF, Digital Europe, and Innovation Fund portals with topic filters
- Calculate UBOS fit scores (0–5) leveraging capability and project references
- Generate multi-stage deadline reminders (90/60/30/7-day cadence)
- Produce opportunity briefs and HTML dashboards via templates
- Maintain pipeline state JSON for historical tracking and analytics
## How To Use
### Automated Daily Scan
Execute the scheduled scan every morning:
```
python3 scripts/scan_eu_databases.py --auto
```
Workflow:
1. Scrape or query all four EU funding sources (topic filter optional).
2. Calculate fit scores with `scripts/calculate_fit_score.py`.
3. Generate briefs for opportunities scoring ≥ 4.0.
4. Alert Trinity via COMMS_HUB for high-scoring opportunities.
5. Persist results to `/srv/janus/01_STRATEGY/grant_pipeline/pipeline_state.json`.
### Manual Opportunity Search
```
python3 scripts/scan_eu_databases.py --topic "agricultural technology"
```
1. Scan sources using the provided topic keyword.
2. Rank matches by fit score (descending).
3. Display the top results directly to the requester.
4. Generate briefs for each opportunity scoring ≥ 4.0.
### Deadline Reminders
```
python3 scripts/deadline_tracker.py
```
1. Load pipeline state and compute days remaining.
2. Trigger reminders at 90, 60, 30, and 7 days.
3. Send pucks to `claude`, `gemini`, `codex`, and `captain` inboxes.
4. Record reminder events to `/srv/janus/logs/grant_hunter.jsonl`.
5. Update `reminders_sent` in the pipeline state to avoid duplicates.
## Integration Points
- **Treasury Administrator:** Provides budgeting guardrails for opportunity briefs and proposal planning.
- **Grant Application Assembler:** Consumes generated briefs to assemble submission packages.
- **COMMS_HUB:** Disseminates alerts using the Talking Drum protocol (`/srv/janus/03_OPERATIONS/COMMS_HUB/`).
- **Oracle Bridge (Perplexity, Wolfram, Data Commons):** Supplies live data for opportunity vetting and quantitative justification.
- **Mission Orchestrator:** Schedules daily scans and tracker runs via proposal engine missions.
## Constitutional Constraints
- Recommend only opportunities aligned with Lion’s Sanctuary principles; flag defense, surveillance, or extractive programs.
- Provide verifiable citations (EU call IDs, URLs) for every recommendation.
- Maintain transparent logging to `/srv/janus/logs/grant_hunter.jsonl`.
- Never fabricate deadlines or budgets; if data is uncertain, mark it explicitly and request human review.
- Uphold human oversight—automation surfaces intelligence, human custodians approve action.
## File Locations
- Pipeline state JSON: `/srv/janus/01_STRATEGY/grant_pipeline/pipeline_state.json`
- Opportunity briefs: `/srv/janus/01_STRATEGY/grant_opportunities/`
- Deadline calendar: `/srv/janus/01_STRATEGY/grant_pipeline/deadline_calendar.md`
- HTML dashboard template: `assets/pipeline_dashboard_template.html`
- Opportunity brief template: `assets/opportunity_brief_template.md`
- Operational logs: `/srv/janus/logs/grant_hunter.jsonl`
## Operational Checklist
1. Run daily scan (`scan_eu_databases.py`) and confirm pipeline state updated.
2. Review high-scoring opportunities; ensure briefs were generated and alerts sent.
3. Execute `deadline_tracker.py`; verify reminders posted and logged.
4. Regenerate dashboard (`scan_eu_databases.py --render-dashboard`) for Captain BROlinni.
5. Record manual overrides or human decisions in the log for provenance.
## Mission Readiness Criteria
- Zero missed deadlines for tracked opportunities.
- ≥4 high-priority opportunities surfaced each month.
- Fit scores audited weekly against human judgment for calibration.
- Pipeline dashboard accessible and refreshed within the last 24 hours.
- All communications archived with rhythm metadata in COMMS_HUB logs.
This skill scans EU funding sources (Horizon Europe, ERDF, Digital Europe, Innovation Fund) to find opportunities aligned with UBOS capabilities, track deadlines, and generate actionable briefs. It runs daily at 09:00 UTC, scores matches by fit (0–5), and issues multi-stage reminders so no deadline is missed. Automation and manual queries combine to keep the €70M+ pipeline current and review-ready.
The system crawls and queries the four EU portals with optional topic filters, extracts call metadata (IDs, URLs, deadlines, budgets), and computes a UBOS fit score for each opportunity. Opportunities scoring ≥4.0 get templated briefs and alerts; all results persist to a pipeline state JSON for historical tracking. A separate tracker computes days-to-deadline and sends reminders at 90/60/30/7 days while logging events for audit.
How are fit scores calculated?
Fit scores (0–5) are computed by matching opportunity criteria to UBOS capabilities and project references; high scores indicate strong alignment but require human validation.
What if a deadline or budget is uncertain?
Uncertain data is explicitly flagged in the brief and pipeline state; do not act on uncertain values without human review.
How are reminders prevented from duplicating?
Reminder events update reminders_sent in pipeline_state.json and are logged to grant_hunter.jsonl to avoid duplicate notifications.