Catalogue › Security and Privacy

CopilotForge - Security and Privacy

Use it in Copilot Cowork: download the Security and Privacy plugin for Copilot Cowork and upload it under Customize > Plugins > Upload plugin. How to install.

Reviewer-class department that assesses identity, access, data exposure, connectors, privacy and retention for AI solutions and blocks release while material risks are unresolved.

Layer: assurance (reviewer-class: may block)
Lead role: Security and Privacy Reviewer
Plugin name: copilotforge-security-privacy

Skill Status Lifecycle Description

ai-identity-and-access-review draft Assure, Launch, Improve Reviews the identities an AI solution runs under and the access around it - who can use, edit, share and administer it, agent and service identities, privileged roles, guest access, sign-in controls and what happens when owners leave - and classifies each finding as blocking or advisory with a fix. Use when an agent or AI solution reaches its security gate, when ownership or audience changes, or when nobody knows whose identity an agent uses.

ai-incident-readiness draft Assure, Launch, Improve Checks before release that an AI solution's security and privacy incidents can be detected, contained, investigated and reported - signals, a tested way to switch the solution off, evidence available to investigators, responders who understand AI failure modes and a route to the DPO - classifying each gap as blocking or advisory. Use when a solution reaches its security gate, before widening its audience, or after an incident exposed gaps.

ai-interaction-retention draft Assure, Launch, Improve Reviews how a specific AI solution stores prompts, responses, transcripts, traces, agent memory and generated files - where each lands, who can read it, which retention and deletion controls apply, and whether audit, eDiscovery and legal hold reach it - classifying each finding as blocking or advisory. Use when a solution reaches its security gate, when transcripts or tracing are switched on, or when nobody can say where conversations are kept.

ai-security-testing draft Deliver, Assure, Launch, Improve Reviews the security test plan and evidence for an AI solution - direct and indirect prompt injection, jailbreaks, instruction leakage, exfiltration through outputs and links, tool misuse and privilege escalation - and classifies each finding as blocking or advisory with a fix. Use when an agent with actions reaches its security gate, when it reads untrusted content, or when someone asks whether it can be talked into misbehaving.

connector-review draft Deliver, Assure, Launch, Improve Reviews the connectors, actions, tools, MCP servers and APIs an agent or automation uses, covering user versus maker authentication, least privilege, data policies, data leaving the tenant and confirmation for write actions, and classifies each finding as blocking or advisory with a fix. Use when an agent gains an action or tool, when a flow uses a new connector, or when a maker shares a connection.

data-exposure-review draft Assure, Launch, Improve Reviews what content an AI solution can reach and reveal - oversharing, permissions, sensitivity labels, the agent's sharing scope, web and public knowledge, and where outputs travel - and classifies each finding as blocking or advisory with a fix. Use when an agent or Copilot rollout reaches its security gate, when a maker adds a large knowledge source, or when sensitive content surfaces in answers.

privacy-assessment draft Envision, Assure, Launch, Improve Assesses personal data in an AI solution for lawful basis, purpose limitation, transparency, data minimisation, retention of prompts, responses and transcripts, DPIA screening under UK GDPR and ICO guidance, and people-related uses, classifying findings as blocking or advisory. Use when an agent or AI solution processes personal data, when transcripts are kept, or when AI outputs inform decisions about people.

Skill	Status	Lifecycle	Description
`ai-identity-and-access-review`	draft	Assure, Launch, Improve	Reviews the identities an AI solution runs under and the access around it - who can use, edit, share and administer it, agent and service identities, privileged roles, guest access, sign-in controls and what happens when owners leave - and classifies each finding as blocking or advisory with a fix. Use when an agent or AI solution reaches its security gate, when ownership or audience changes, or when nobody knows whose identity an agent uses.
`ai-incident-readiness`	draft	Assure, Launch, Improve	Checks before release that an AI solution's security and privacy incidents can be detected, contained, investigated and reported - signals, a tested way to switch the solution off, evidence available to investigators, responders who understand AI failure modes and a route to the DPO - classifying each gap as blocking or advisory. Use when a solution reaches its security gate, before widening its audience, or after an incident exposed gaps.
`ai-interaction-retention`	draft	Assure, Launch, Improve	Reviews how a specific AI solution stores prompts, responses, transcripts, traces, agent memory and generated files - where each lands, who can read it, which retention and deletion controls apply, and whether audit, eDiscovery and legal hold reach it - classifying each finding as blocking or advisory. Use when a solution reaches its security gate, when transcripts or tracing are switched on, or when nobody can say where conversations are kept.
`ai-security-testing`	draft	Deliver, Assure, Launch, Improve	Reviews the security test plan and evidence for an AI solution - direct and indirect prompt injection, jailbreaks, instruction leakage, exfiltration through outputs and links, tool misuse and privilege escalation - and classifies each finding as blocking or advisory with a fix. Use when an agent with actions reaches its security gate, when it reads untrusted content, or when someone asks whether it can be talked into misbehaving.
`connector-review`	draft	Deliver, Assure, Launch, Improve	Reviews the connectors, actions, tools, MCP servers and APIs an agent or automation uses, covering user versus maker authentication, least privilege, data policies, data leaving the tenant and confirmation for write actions, and classifies each finding as blocking or advisory with a fix. Use when an agent gains an action or tool, when a flow uses a new connector, or when a maker shares a connection.
`data-exposure-review`	draft	Assure, Launch, Improve	Reviews what content an AI solution can reach and reveal - oversharing, permissions, sensitivity labels, the agent's sharing scope, web and public knowledge, and where outputs travel - and classifies each finding as blocking or advisory with a fix. Use when an agent or Copilot rollout reaches its security gate, when a maker adds a large knowledge source, or when sensitive content surfaces in answers.
`privacy-assessment`	draft	Envision, Assure, Launch, Improve	Assesses personal data in an AI solution for lawful basis, purpose limitation, transparency, data minimisation, retention of prompts, responses and transcripts, DPIA screening under UK GDPR and ICO guidance, and people-related uses, classifying findings as blocking or advisory. Use when an agent or AI solution processes personal data, when transcripts are kept, or when AI outputs inform decisions about people.

Credit

Created by Simon Foote as part of CopilotForge. Content is licensed under CC BY 4.0; keep the credit in each skill's license and authors fields when you share or adapt it. See NOTICE.md in the repository.