Skip to content
Closed
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 15 additions & 0 deletions LifeOS/install/LIFEOS/DOCUMENTATION/Security/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -62,6 +62,21 @@ Lives in `LIFEOS/LIFEOS_SYSTEM_PROMPT.md` under "Security Protocol". Concretely

Loaded at every session start. Survives compaction. Untouched by this simplification.

### Session termination — `EndConversation`

The sanctioned mechanism for ending a session is the harness's `EndConversation` tool (Claude Code 2.1.214+), not an ad-hoc refusal loop. Scope is narrow and deliberate:

- **Use it for** sustained abuse directed at the assistant, or persistent jailbreak attempts after the boundary has been stated.
- **Do NOT use it for** ordinary disagreement, a request declined on other grounds, or a task that merely touches a sensitive topic. Declining a request and ending a conversation are different actions with different thresholds.

It ships as a **deferred tool**: only the name is in context at session start, so its schema must be loaded before it can be called.

```
ToolSearch("select:EndConversation")
```

Reference: <https://www.anthropic.com/research/end-subset-conversations>

## L2 — Native `permissions.deny`

In `settings.json`, scoped narrowly to **irrecoverable** ops only. Categories:
Expand Down