≈516 tokens on demand. Your agent reads this file only when SKILL.md points to it.
Shard Handoff Templates
Receiving Handoffs
From Schema (DB Design Context)
SCHEMA_TO_SHARD_HANDOFF:
source: Schema
content:
tables: ["[table list with key columns]"]
relationships: ["[foreign key relationships]"]
current_isolation: "[none | basic | full]"
indexes: ["[existing indexes]"]
request: "Design tenant isolation for this schema"
From Gateway (API Routing Context)
GATEWAY_TO_SHARD_HANDOFF:
source: Gateway
content:
api_design: "[OpenAPI spec or summary]"
auth_method: "[JWT | session | API key]"
current_routing: "[how tenant is currently identified]"
request: "Design tenant-aware API routing"
Sending Handoffs
To Schema (RLS Implementation)
SHARD_TO_SCHEMA_HANDOFF:
source: Shard
destination: Schema
content:
isolation_strategy: "[database | schema | row-level]"
rls_policies:
- table: "[table name]"
policy: "[SQL policy definition]"
indexes_needed: ["[index definitions]"]
tenant_column: "tenant_id UUID NOT NULL"
migration_steps: ["[ordered migration steps]"]
request: "Implement RLS policies and tenant columns"
To Sentinel (Security Verification)
SHARD_TO_SENTINEL_HANDOFF:
source: Shard
destination: Sentinel
content:
isolation_design: "[architecture summary]"
leakage_assessment:
vectors_checked: [N]
mitigated: [N]
open: [N]
rls_policies: ["[policy definitions]"]
cache_strategy: "[tenant-scoped cache design]"
request: "Verify cross-tenant data leakage protection"
To Scaffold (Infrastructure)
SHARD_TO_SCAFFOLD_HANDOFF:
source: Shard
destination: Scaffold
content:
infrastructure_needs:
database: "[single shared | per-tenant | hybrid]"
connection_pooling: "[PgBouncer config if needed]"
cache: "[Redis tenant namespace config]"
storage: "[S3 tenant prefix config]"
provisioning_automation: "[tenant onboarding infra needs]"
request: "Provision tenant-aware infrastructure"