feat: add OPENCLAUDE_DISABLE_TOOL_REMINDERS env var to suppress hidden tool-output reminders (#837)

Gates three injection sites behind OPENCLAUDE_DISABLE_TOOL_REMINDERS:
- FileReadTool cyber-risk mitigation reminder (appended to every Read
  result when the model is not in MITIGATION_EXEMPT_MODELS)
- todo_reminder attachment for TodoWrite usage
- task_reminder attachment for TaskCreate/TaskUpdate usage

All three reminders are model-only side-channel instructions the user
cannot see today. Users who want full transparency over what the model
receives can now opt out without patching dist/cli.mjs on every upgrade.

Default behavior is unchanged when the flag is unset.

Closes #809
This commit is contained in:
0xfandom
2026-04-22 23:07:02 +05:30
committed by GitHub
parent 23e8cfbd5b
commit 28de94df5d
3 changed files with 15 additions and 0 deletions

View File

@@ -733,6 +733,9 @@ export const CYBER_RISK_MITIGATION_REMINDER =
const MITIGATION_EXEMPT_MODELS = new Set(['claude-opus-4-6'])
function shouldIncludeFileReadMitigation(): boolean {
if (isEnvTruthy(process.env.OPENCLAUDE_DISABLE_TOOL_REMINDERS)) {
return false
}
const shortName = getCanonicalName(getMainLoopModel())
return !MITIGATION_EXEMPT_MODELS.has(shortName)
}