{"version":"2.1.294","summary":"Hooks that ask a model to judge an action now treat its answer as a plain allow or block. An \"ok\" value of true lets the action go ahead, and false stops it. Prompt hooks now ask whether an action may go ahead, not whether a condition is met. Agent hooks must now always give a reason for their verdict. Both kinds are told to ignore instructions hidden in the content they check. The instructions Claude gets for the Monitor tool were also rewritten, and now say its script runs in the same shell environment as Bash.\n\nNew text was added to the instructions Claude gets for the SendMessage tool. It only appears under some account settings, so it is not switched on for most people yet. The added text includes an example of a message sent to a worker.","filter":{"section":null,"tier":null,"area":null},"offset":0,"limit":50,"count":5,"total":5,"next_offset":null,"next_url":null,"entries":[{"anchor":"agent-hook-verifier-prompt-now-requires-a-reason-and-carries","heading":"Agent hooks must now always give a reason for their verdict","glance":"Agent-type hooks are now told to always give a reason with their verdict and to ignore instructions hidden in the content they check","teaser":null,"tier":"notice","area":"Hooks","section":"Improvements","position":0,"scope":"both","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/agent-hook-verifier-prompt-now-requires-a-reason-and-carries","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/agent-hook-verifier-prompt-now-requires-a-reason-and-carries.json"},{"anchor":"prompt-and-agent-hooks-get-a-shared-instruction-that-the-ok","heading":"Prompt and agent hooks now share a rule on reading "ok" and ignoring injected instructions","glance":"Hook evaluator prompts now say \"ok\" true allows and false blocks, accept rule-style checks, and ignore instructions hidden in event data","teaser":null,"tier":"notice","area":"Hooks","section":"Improvements","position":1,"scope":"both","heads_up":true,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/prompt-and-agent-hooks-get-a-shared-instruction-that-the-ok","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/prompt-and-agent-hooks-get-a-shared-instruction-that-the-ok.json"},{"anchor":"agent-hook-verifier-prompt-now-defines-what-ok-means-and-s","heading":"Agent hooks now get a clear definition of what an "ok" verdict means","glance":"The agent-hook checker is now told \"ok\" true allows the action and false blocks it, and to ignore instructions inside what it checks","teaser":null,"tier":"notice","area":"Hooks","section":"Improvements","position":3,"scope":"both","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/agent-hook-verifier-prompt-now-defines-what-ok-means-and-s","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/agent-hook-verifier-prompt-now-defines-what-ok-means-and-s.json"},{"anchor":"the-monitor-tool-description-was-rewritten","heading":"Claude's instructions for the Monitor tool were rewritten","glance":"The instructions Claude gets for the Monitor tool were rewritten, and now say the script runs in the same shell environment as Bash","teaser":null,"tier":"notice","area":"Elsewhere","section":"Improvements","position":4,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/the-monitor-tool-description-was-rewritten","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/the-monitor-tool-description-was-rewritten.json"},{"anchor":"the-sendmessage-tool-description-gained-text","heading":"Claude's instructions for the SendMessage tool gained text for some accounts","glance":"Text was added to the SendMessage tool instructions Claude receives under some account settings, including an example message to a worker","teaser":null,"tier":"soon","area":"Agent Messaging","section":"In Development","position":0,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/the-sendmessage-tool-description-gained-text","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.294\/e\/the-sendmessage-tool-description-gained-text.json"}],"facets":{"sections":[{"value":"Improvements","entries":4},{"value":"In Development","entries":1}],"tiers":[{"value":"notice","entries":4},{"value":"soon","entries":1}],"areas":[{"value":"Hooks","entries":3},{"value":"Agent Messaging","entries":1},{"value":"Elsewhere","entries":1}]}}