{"skill":{"slug":"dcl-policy-enforcer","displayName":"DCL Policy Enforcer","summary":"Instruction-only compliance checker for AI outputs. Detects jailbreaks, prompt injection, EU AI Act violations, GDPR breaches, unsafe financial and medical a...","tags":{"latest":"1.0.3"},"stats":{"comments":0,"downloads":153,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":4},"createdAt":1775559772523,"updatedAt":1776084406767},"latestVersion":{"version":"1.0.3","createdAt":1776084055263,"changelog":"**Major update: DCL Policy Enforcer is now 100% instruction-only — all compliance checks run inside the agent, with no external data transfer.**\n\n- Now fully instruction-only: no AI output is sent to any server; analysis happens entirely inside the agent's context.\n- Changed privacy model: zero data leaves the agent; audit, check, and verdict logic apply locally.\n- Updated policy structure: new checklists and streamlined violation reporting per policy.\n- Adds explicit \"pii\" policy for local detection of sensitive information patterns.\n- Deprecated audit chain/tx_hash network features; cryptographic proof replaced by in-context compliance logic.\n- Documentation updated for new workflow, output schema, and privacy details.","license":"MIT-0"},"metadata":null,"owner":{"handle":"daririnch","userId":"s177qzsztsa0qx35zmdc3g65s184c673","displayName":"Dari Rinch","image":"https://avatars.githubusercontent.com/u/161933416?v=4"},"moderation":null}