{"version":"2.1.292","anchor":"hook-systemmessage-and-additionalcontext-now-wrapped-before","canonical_anchor":"hook-systemmessage-and-additionalcontext-now-wrapped-before","heading":"Hook messages and added context are processed before reaching the model","tier":"notice","area":"Hooks","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.292\/e\/hook-systemmessage-and-additionalcontext-now-wrapped-before","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.292","markdown":"### Hook messages and added context are processed before reaching the model\n\nText from a hook's systemMessage and additionalContext now goes through an extra processing step before the model sees it\n\n**Unclear.** What the extra processing step does to hook text is not stated.\n\n**What**\n\nHooks (your own scripts that Claude Code runs at set points) can hand text back to the model through `systemMessage` and `hookSpecificOutput.additionalContext`. That text now goes through an extra processing step before it is added to the conversation. Before, it was added exactly as the hook returned it.\n\n**Why**\n\nThe way hook text appears to the model may be different, so check any hook that depends on its exact wording.\n\n- Area: Hooks\n- Names: `systemMessage`, `hookSpecificOutput.additionalContext`\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}