Andara Rag Search

Search the Andara Ionic RAG knowledge base (3,800+ records) for business intel, research, products, team, meetings, and any indexed content.

Security Scan

The skill's purpose (searching a Postgres-backed RAG) is plausible, but the SKILL.md expects a DATABASE_URL env var and the psql binary while the skill metadata declares no required environment variables or binaries — this mismatch and the need for direct DB access are concerning.

Purpose & Capability

The skill claims to search a Postgres-backed RAG which legitimately requires DB access. However, the metadata lists no required env vars or binaries even though the instructions require psql and a DATABASE_URL. That mismatch is disproportionate to the declared requirements.

Instruction Scope

SKILL.md explicitly instructs the agent to run psql queries against a knowledge_base and other tables. It also references truncation and read-only rules, but the agent would need direct DB credentials to run these commands. The instructions give the agent ability to run arbitrary SQL via psql; there is no explicit technical constraint preventing broader queries beyond the 'read-only' guideline.

Install Mechanism

Instruction-only skill with no install spec or downloaded code. This minimizes disk persistence and supply-chain risk.

Credentials

The instructions rely on an environment variable (DATABASE_URL) that is not declared in requires.env or primary credential. That implies the skill expects database credentials but does not declare them, which is a transparency/permission mismatch. The tables described include sensitive fields (emails, orders, equity_percent), so DB credentials should be least-privileged and explicitly requested.

Persistence & Privilege

always is false and there are no install-time changes or config paths. The skill does not request persistent or elevated platform privileges in its metadata.

What to consider before installing

Before enabling this skill, be aware it requires direct read access to your Postgres database (it expects DATABASE_URL and the psql client). The skill metadata does not declare these needs — ask the publisher to update requires.env and required binaries. If you proceed, provide a dedicated, read-only database user scoped to only the tables/columns needed (avoid production superuser credentials), consider using a read-replica, and review the exact SQL the agent will be allowed to run. Note that while SKILL.md says 'read-only', the agent runs psql and could run other SQL unless you enforce restrictions at the DB/user level. If you cannot provide a tightly-scoped read-only DB user and cannot verify the skill author, treat this skill as risky.

Andara RAG Knowledge Search

Use this skill when you need to find information about:

• Team members, roles, responsibilities
• Products, pricing, supplier details
• Scientific research, water science, bioelectricity
• Business structure, equity, governance
• Meeting notes, action items, decisions
• CMS pages, content, topic clusters
• Orders, customers, revenue data
• Any business intelligence question

How to Use

Run a PostgreSQL query against the knowledge_base table using the bash tool:

psql "$DATABASE_URL" \
  -c "SELECT content::text, source, data_type FROM knowledge_base WHERE content::text ILIKE '%SEARCH_TERM%' LIMIT 5;"

Replace SEARCH_TERM with the relevant keyword(s).

Available Tables

Core Data

Table	Description	Key Columns
knowledge_base	RAG chunks (3,800+)	content, source, data_type
team_members	Team roster (6 active)	name, role, title, email, department, equity_percent
team_meetings	Meeting notes (5)	title, date, summary, key_insights, decisions
meeting_action_items	Action items (32)	title, assignee, status, priority, due_date
team_goals	Company goals (4)	title, status, target_date, progress_percent

CMS & Commerce

Table	Description	Key Columns
pages	CMS pages (155)	slug, title, content, zone, cluster_id, status
products	Products (2)	name, price_cents, description, sku
orders	Orders (11)	total, status, customer_name, created_at
customers	Customers (10)	name, email, created_at
clusters	Topic clusters (20)	name, slug, description

Intelligence

Table	Description
rag_memory_objects	Learned lessons & policies
science_articles	Scientific content
newsletter_subscribers	Email subscribers

Example Queries

Find team member info

psql "$DATABASE_URL" \
  -c "SELECT name, title, department, equity_percent FROM team_members WHERE is_active = true;"

Search knowledge base

psql "$DATABASE_URL" \
  -c "SELECT LEFT(content::text, 300) as content, source FROM knowledge_base WHERE content::text ILIKE '%chris%' LIMIT 5;"

Get CMS page content

psql "$DATABASE_URL" \
  -c "SELECT slug, title, zone FROM pages WHERE status = 'published' ORDER BY slug LIMIT 20;"

Get revenue summary

psql "$DATABASE_URL" \
  -c "SELECT COUNT(*) as orders, SUM(total)/100.0 as revenue_eur, AVG(total)/100.0 as avg_order FROM orders;"

Get meeting action items by person

psql "$DATABASE_URL" \
  -c "SELECT title, status, priority, due_date FROM meeting_action_items WHERE assignee ILIKE '%chris%';"

Get page content by slug

psql "$DATABASE_URL" \
  -c "SELECT title, LEFT(content, 500) as content_preview FROM pages WHERE slug = '/ion/overview';"

Rules

• Always use LEFT(content::text, 300) to truncate long content fields (content is jsonb, must cast to text)
• Default LIMIT to 5-10 results to keep responses concise
• Use ILIKE for case-insensitive text searches
• Never INSERT, UPDATE, or DELETE — read-only access only
• For questions about the website structure, query the pages table
• For scientific questions, search knowledge_base WHERE data_type = 'research' OR search science_articles