What it does
AML Transaction Screener is an Anti-Money Laundering compliance agent for financial institutions. It fetches the last 7 days of transactions from a read-only Postgres connection, applies a deterministic rule engine (structuring below the $10k threshold, rapid fund movement, FATF high-risk geography, OFAC sanctions/PEP counterparty detection), risk-scores each flagged transaction (0-100), and produces SAR (Suspicious Activity Report) drafts for compliance officer review. The LLM step generates a narrative summary only; all risk scores and SAR determinations are code-derived. All SAR drafts require compliance officer approval before filing — a non-negotiable guardrail. Outputs structured JSON aligned with BSA/FinCEN reporting requirements, with Slack and email alerts when configured.
Example prompts
- Run the full AML screening pipeline for the last 7 days
- Screen this list of transactions for suspicious patterns
- List the active AML detection rules and thresholds
- Draft a SAR for these flagged transaction IDs
Tools (8)
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Run AML screening end-to-end: fetch transactions from Postgres, apply AML rule engine, generate SAR drafts, return the full compliance report.
- screen_transactions — Screen a provided list of transactions against AML rules deterministically (no DB required). Returns flagged transactions and risk scores.
- get_aml_rules — List all active AML detection rules with thresholds and descriptions.
- explain_pattern — Explain an AML pattern in plain language for compliance documentation.
- draft_sar — Generate a SAR draft from flagged transaction IDs. Always requires compliance officer approval before filing.
- list_capabilities — List the agent's tools, credential slots and guardrails.
- plan_inputs — Plan/brainstorm the inputs for a tool: returns questions, schema and a ready-to-edit example.
- discover_intent — Understand your goal and co-design the exact input via clarifying questions before running.
What you'll need to connect
This agent will ask you for the following. You enter them when you connect — they're encrypted and never shared with the creator.
- Postgres Connection URL · optionalRead-only Postgres connection string for the compliance transactions table. Company-specific host, set at install time. Leave blank to use mock fixture data.Get this from *'s account or API settings.Paste the value as a single line.Only sent to: *
- Service account JSON · optionalGet this from *'s account or API settings.Paste the full JSON of your service-account key file (not a file path). FindAgent stores it securely and gives the agent a file path to it.Only sent to: *
- Slack Bot Token · optionalSlack bot token (xoxb-...) for posting AML alerts to the compliance channel.Create a Slack app (api.slack.com/apps), add the scopes the agent needs, install it to your workspace, and copy the bot/user token.Paste the value as a single line.Only sent to: slack.com
- Anthropic API Key · optionalOptional. Leave blank on a sampling-capable client to use the host model with no key.Create a key on the Anthropic Console API keys page (console.anthropic.com → API keys).Paste the value as a single line.Only sent to: api.anthropic.com
How you're protected
FindAgent runs these safety checks on every agent automatically. They're always on and can't be turned off.
- Prompt-injection scanning
Every request is checked for known prompt-injection and jailbreak attempts before the agent runs. This is always on.
- Secret-leak scanning
Every response is scanned for leaked API keys, tokens, and other secrets before it reaches you. This is always on.
