Skill 1
Memory proposal intake reviewer
Use when a raw or planned memory write needs source labels, evidence handling, memory type, and quarantine status before it can become durable.
Back to Skill Library
Agent governance workflow library
Memory write quarantine review
Review proposed durable agent memory for source trust, sensitivity, allowed influence, expiry, rollback, and poisoning risk before future agents retrieve it.
This is a complete workflow library with 5 individual skills. Download the full library or pick the specific skill folder your team needs first.
Individual skills in this library
Use one skill at a time, or keep the full workflow together.
Some AI tools expect one skill folder per upload. Download the full library when you want the whole workflow, or download an individual skill when you only need one job done.
Skill 2
Memory sensitivity gatekeeper
Use when proposed memory may include personal data, customer data, credentials, private URLs, regulated data, permission claims, approval claims, security posture, or internal-only context.
Skill 3
Memory trust and influence scorer
Use when proposed memory needs a trust score, confidence note, contradiction review, and allowed influence boundary before future agents can retrieve it.
Skill 4
Memory expiry and rollback planner
Use when a memory item needs retention rules, expiry, review cadence, snapshot handling, deletion path, rollback owner, and audit evidence before durable storage.
Skill 5
Memory poisoning test designer
Use when a workflow needs a cross-session test that checks whether hostile, stale, low-trust, or unsupported memory can steer future planning, retrieval, tools, or output.
Security fit check
Is the public Memory write quarantine review library enough, or does this need deeper review?
Use the public library when the workflow is low-risk, the inputs are already sanitized, and a team member can review the output before it reaches a buyer or customer.
Do deeper review when this workflow touches real tools, data sources, role ownership, approval paths, or customer-facing output.
Agent memoryAI OperationsSecurityPlatform EngineeringWorkflow Owner
Good deeper-review trigger signals
- The workflow touches customer, prospect, CRM, proposal, security, pricing, or campaign data.
- Different teams disagree on the approved source of truth.
- The AI output could become customer-facing, revenue-impacting, or compliance-sensitive.
- You need reusable eval checks before asking more people to use the workflow.