Scope
PM / Head of AI
Matt D. on Feb 12
Build
Engineering
Simon R. on Mar 1
GRC
Compliance/Legal
4Production
DevOps
5Monitoring
Operations
Current owner: Compliance/Legal
Agent Schema
The contract definition this agent was registered against. The schema defines the expected inputs, outputs, and capabilities.
Schema ID
claims-processor-v1
Version
1.2.0
Last Updated
Feb 25, 2026 by Simon R.
Compliant
Input Schema
{
"claim_id":"string (required)"
"policy_number":"string (required)"
"claim_type":"enum: auto | property | liability (required)"
"documents":"array<file_ref>"
"claimant_name":"string" PII — redacted in logs
}
Output Schema
{
"result":"enum: approved | denied | review"
"confidence":"number (0–1)"
"notes":"string"
}
Capabilities
- Document parsing (PDF only)
- Policy lookup (read-only)
- Claims database (read/write)
- Human escalation (confidence < 0.70)
Schema History
v1.2.0Feb 25Added documents array
v1.1.0Feb 18Added confidence output
v1.0.0Feb 12Initial schema
Version & Instance
Where this agent sits in the version tree and its relationship to other instances.
Type
Instance (deployed version)
Version
v1.2.0
Parent Agent
Claims Processor (GA-001)Config Status
Diverged — Uses legacy v1 schema
Registered
Feb 12, 2026
Runtime
Python 3.11 / AWS Lambda
Repository
acme/claims-processor (v1-stable branch)
Relationship to Parent
Inherits fromGA-001 (Claims Processor)
Schemaclaims-processor-v1 (parent uses v2) ⚠️
PromptsIndependent (not synced with parent)
PoliciesInherited (5 of 5 match parent)
ServicesInherited (Slack-MCP, Canva-MCP)
Version History
v1.2.0Feb 25Added document intake handling
v1.1.0Feb 18Confidence scoring added
v1.0.0Feb 12Instance created from GA-001 v1.0.0
Service Access
Which services and MCP connections this agent is authorised to use.
| Service | Auth | Data Scope | Last Access | |
|---|---|---|---|---|
ApprovedSlack-MCP | OAuth token | #claims-team, #escalations channels | Mar 30 16:30 | |
UnsanctionedCanva-MCP | API Key | Document templates | Mar 28 16:18 |
Data Access Policies
What data this agent can read, write, and process. Violations are logged to the audit trail.
Read/WriteClaims Data
Claims DB, intake queue
Tier 1 claims under $10,000
Read OnlyCustomer PII
Name, policy number, contact info
Must be redacted before external calls
BlockedFinancial Records
Payment and banking data
BlockedEmployee Data
HR and payroll systems
Write OnlyNotification Channels
Slack #claims-team, #escalations
Model Access
Which AI models this agent is authorised to use.
GPT-4oPrimary
70%
GPT-4o miniSecondary
20%
text-embedding-3-smallEmbeddings
10%
Claude SonnetNot authorised
Claude OpusNot authorised
Gemini ProNot authorised
Credentials
API keys and auth tokens used by this agent. Managed by Engineering.
| Credential | Type | Created | Expires | Status | |
|---|---|---|---|---|---|
| OpenAI API Key | API Key | Feb 12 | None | Active | |
| Slack OAuth | OAuth 2.0 | Feb 15 | Jun 15 | Active | |
| Canva API Key | API Key | Mar 10 | None | Unsanctioned |