@misterhuydo/cairn-mcp 1.31.0 → 1.33.0

package/README.md

@@ -374,23 +374,43 @@ is least reason to speak. Bytes survive now only as a fallback for when no usage
 record is reachable, and when that happens the line says so instead of dressing a
 file size up as a token count.
 
-**It only speaks at 90% of the window and above.** The window size is not observable
-— no hook input carries it, and the transcript records usage but never a limit — so
-cairn infers the smallest window that could already hold what it has measured.
-Having *seen* 482k tokens proves the window is bigger than 200k, and that is the one
-direction the inference needs to be sound in. A fixed absolute threshold cannot
-serve both sizes: 120,000 tokens is 60% of a 200k window but 12% of a 1M one, which
-is why v1.20.0 announced a stopping point on a session sitting at just over half its
-context. Override with `CAIRN_CONTEXT_HINT_TOKENS` if you want a different bar.
-
-The one case the inference gets wrong: a 1M-window session between roughly 180k and
-196k looks like a nearly-full 200k one, so it may speak once. It corrects itself as
-soon as usage passes 200k, and the no-repeat rule means it says it once, not every
-turn.
+**There is no threshold, and since v1.33.0 there is deliberately no attempt at one.**
+Cairn reports the measured count on every prompt and the *agent* decides whether to
+stop, weighing what is left against how big the work you just asked for is going to be.
+
+The reason is that a threshold needs a window, and the window is the one thing cairn
+cannot see. No hook input carries it, the transcript records usage but never a limit,
+and the model id is no help either: a `claude-opus-5[1m]` session writes the base
+`claude-opus-5` into both the transcript's `message.model` and `lastModelUsage` in
+`~/.claude.json`. Three versions tried anyway, each failing differently:
+
+| | approach | how it failed |
+|---|---|---|
+| v1.20.0 | fixed 120k tokens | 60% of a 200k window but 12% of a 1M one, so it fired a tenth of the way into a 1M session |
+| v1.31.0 | infer the smallest window that fits the measurement | sound as a lower bound only, so below 196k it assumed 200k — reported **96%** where `/context` said **19%** |
+| v1.32.0 | never guess downward; `CAIRN_CONTEXT_WINDOW` states the truth | nothing could ever set it. `install-hooks` cannot know the window either, so there was no correct value to write — unwritable by construction, and the feature went silently dead |
+
+v1.31.0's false-alarm band could never have been tuned away: any firing point for a
+200k window is a point a 1M session also passes through, so no number serves both.
+
+The agent, though, *does* know its window — it is stated in its own system prompt —
+and it is the only party that knows whether the next task is a one-line answer or a
+refactor across thirty files. So cairn hands over what it actually measured and the
+rules for acting on it, and never issues the verdict itself. It is the same division
+of labour as the handoff: cairn demands the judgement and carries it, never computes
+it.
+
+That also removes the rate-limiting. The old block was a *verdict*, so repeating it
+was nagging and it had to be shown once per situation; this is a *measurement*, and a
+budget you are only shown once it is nearly spent cannot inform the decision to start
+something. It rides every prompt.
+
+`CAIRN_CONTEXT_HINT_TOKENS` no longer gates anything, but if you set it, cairn tells
+the agent when you have passed it rather than ignoring you.
 
 Deliberately not built: no full transcript walk (the tail only; a full walk in this
 hot path once froze every terminal in the cockpit for 4.3 seconds), no counting
-subagent spend as main-thread context, no repeating the prompt when nothing changed,
+subagent spend as main-thread context, no cairn-side verdict about when to stop,
 and no auto-clearing, auto-compacting, or filling the field in on your behalf. Same
 shape as completed-phase detection: detect, surface, confirm with the user, never apply.