Catches the mistakes coding agents repeat before they merge: a pull-request gate with no model call and no API key.
Short answer: Every DB read/write filters by authenticated workspace scope. One query without the tenant filter leaks every tenant’s rows.
Group: Security and tenancy. Status: stable. Generated from the rule definitions; do not edit by hand.
Playground: https://brevio.pro/tools/agent-guard-playground?rule=R2.2
Tenant-scoped data access: every DB read/write filters by authenticated workspace scope.
One query without the tenant filter leaks every tenant’s rows.
db.select().from(orders).where(eq(orders.id, id))
db.select().from(orders).where(and(eq(orders.id, id), eq(orders.tenantId, session.tenantId)))
main, in a git checkout (dev_rules.scan_added_lines); brevio-core scan does not run it.dev_rules.slice_micro_rules).A person can silence one finding with brevio-core: ignore R2.2 -- reason on the line or the line above. Without the reason it does not count. An agent cannot silence it: the agent hook and the MCP tools never honour the comment, and a pull request that adds one is reported until a reviewer allows it. Fixing the code is still the better answer.
What the scanner already leaves alone: Only API code under apps/api/ is checked, and only a filter the change removes.
A scan reads only the lines you add, so a line that was already in the file is not flagged.
To change what the rule flags, edit its definition in dev_rules.py or code_rules/__init__.py. That is a code change and is reviewed like one.