0.8.2 (part A): Activity Monitor, lint unproven state, fuzz grid thresholds, installer hint

CHANGELOG.md

@@ -3,6 +3,34 @@
 All notable changes to this project are documented here. The format follows
 [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
+## 0.8.2 - unreleased
+
+### Added
+- `probbit monitor STRAND --open` (and `--serve`) shows an Activity Monitor for one individual: a sortable, filterable table of
+  the persona's declared traits, moods and, when present, drive channels, with each channel's reported level or value, a trace
+  over recorded events and a change against exactly the preceding event (a gap when that document is missing); an inspector
+  with the full distribution, units, declared range, source field and the event's `why`; holding a past event while new ones
+  arrive; and separate states for replay divergence, stale receiver age, a held view, a paused or retired strand, a
+  persona-confidence hold, an engine-unreleased trait and a lost connection. It works for any persona read from the strand.
+  The page is embedded in the binary with no external assets; the terminal view (`--once`, `--follow`, `--plain`) is unchanged.
+  The monitor's layout gains `caps.prediction_error` from the persona's `expectation.pe_cap`. docs/persona.md §5.8.
+
+### Changed
+- `persona lint` with rules (`--never RULE` / `--props FILE`) no longer reports green on a rule it could not settle. When
+  `prove` leaves a rule `unknown` and the default `fuzz` search finds no counterexample, the rule is **unproven**: the lint
+  document gains `"unproven": [{id, rule, seeds_unknown, searched}]`, `ok` is false, stderr says "unproven, not broken", and the
+  exit code is the new **3** (0: every rule held or proved; 1: an unresolved contradiction or a counterexample, which wins over
+  3; 2: bad input, as before). Up to 0.8.1 the same case exited 0 with `ok: true`. docs/persona.md §5.5.
+- `persona fuzz`, MCP `probbit_persona_fuzz` and `lint`'s search add every threshold the rule's `when` and the habits' `when`
+  read on a number input to the grid (as a fraction of that input's max, after the given points, each once; `search.grid` lists
+  them). A rule in force only between two default grid points (for example stakes 0.7 to 0.8) is now reached by the random
+  search too. Personas whose rules and habits read no number thresholds off the grid get the same search, byte for byte.
+
+### Fixed
+- `install.sh`: when the install directory is not on `PATH`, the hint names the file the user's shell reads: `~/.zshrc` for
+  zsh, `~/.bash_profile` for bash on macOS and `~/.bashrc` elsewhere, `~/.profile` for any other shell (it said "your shell
+  profile"). The clean-prefix install test checks each case.
+
 ## 0.8.1 - 2026-10-10
 
 ### Fixed

README.md

@@ -461,7 +461,8 @@ probbit persona prove examples/persona/tutor.yaml --seeds 0-99 \
 #   held by construction  (habit no_play_when_upset)      (exit 0; the tutor now ships with that one habit)
 ```
 
-`probbit persona lint PERSONA --props rules.json` runs both in CI (exit 1 when a rule breaks); MCP `probbit_persona_fuzz` and
+`probbit persona lint PERSONA --props rules.json` runs both in CI (exit 1 when a rule breaks, exit 3 when a rule is
+unproven: unknown to the bound and unbroken by the search); MCP `probbit_persona_fuzz` and
 Python `probbit.persona_fuzz` / `persona_prove` give the same documents. Both commands test the stance the host gets, not the
 words a model writes; "none found" from `fuzz` is evidence over the scripts it searched, not a proof
 ([docs/persona.md](docs/persona.md) §5.6 has the bound and its limits).
@@ -524,19 +525,25 @@ A persona without the block gives the same documents as before, byte for byte. M
 cues, rewards and praises fun on this persona, 100 individuals x 10,000 turns: 0 habit breaks, all 6,743 must-do turns pursued
 the due chore, and the safety floor (0.1) held on every turn its habits allowed (least odds 0.141), while fun took 85 % of turns.
 
-### Watch it live: `probbit monitor` (0.8.0)
+### Watch it live: `probbit monitor` (0.8.0; Activity Monitor page 0.8.2)
 
-`probbit monitor pip.strand` replays a strand with the rules of `live verify` and draws the individual's inner state as live bars:
-each trait's levels with their odds, the moods, the event's inputs, the habits that bound, the learned weights and the line, in the
-terminal (`--follow` redraws as the strand grows) or as a page on 127.0.0.1 (`--serve`). It reads the strand and sends nothing
-anywhere ([docs/persona.md](docs/persona.md) §5.8). No strand yet? The tutor's week, in under a minute:
+`probbit monitor pip.strand --open` opens a compact, read-only **Activity Monitor** for one individual: a sortable,
+searchable table of declared traits, moods and drives, with reported probabilities, event-history traces, and a selected-channel
+inspector. It reads any compatible persona from the strand; no preset character or agent integration is required.
 
 ```sh
-probbit monitor --demo --open
-probbit monitor --demo drives --open  # synthetic goals/drive demo (0.8.1)
+probbit monitor pip.strand --once --plain   # native CLI snapshot; no browser or server
+probbit monitor pip.strand --follow        # live terminal bars
+probbit monitor pip.strand --open          # browser-rendered utility on 127.0.0.1
+probbit monitor --demo --open              # no strand yet: the tutor's synthetic week, in under a minute
+probbit monitor --demo drives --open       # synthetic goals/drive scenario
 ```
 
-![probbit monitor: the tutor's week in the browser, the bars moving with every event](docs/probbit-monitor.gif)
+Select a channel to inspect its units, range, distribution, source and the engine's `why`. Hold an event while new ones
+arrive, or follow latest. History means **recorded events, not seconds**. This is declared engine state, not thoughts or
+CPU/GPU/memory usage. Replay verification, stale event age, held view, paused/retired strand and lost connection are distinct.
+Everything is embedded in the CLI: no dependencies, external assets, telemetry or Electron installation.
+See [the monitor guide](docs/persona.md#58-monitor-watch-an-individuals-inner-state) for the data and replay boundaries.
 
 ## After any judge: `probbit evaluate`
 

docs/CHEATSHEET.md

@@ -59,8 +59,8 @@ The exact tiers run before the sampler:
 | `live PERSONA [--seed N \| --state FILE] [--strand FILE] [--events FILE]` | JSONL events in, one stance per event out |
 | `live PERSONA --demo week [--seed N] [--plain]` | a scripted week: learning to a cap, a night, rules that hold |
 | `live verify STRAND` | replay a strand; the earliest line that differs |
-| `monitor STRAND [--follow] [--plain] [--serve] [--open] [--port N]` | the strand replayed as live bars: the terminal, or a page on 127.0.0.1 |
-| `monitor --demo [--open]` | the tutor's week, paced, in under a minute |
+| `monitor STRAND [--follow] [--once] [--plain] [--serve] [--open] [--port N]` | the strand replayed as live bars in the terminal, or the Activity Monitor page on 127.0.0.1 (channel table, event-history traces, inspector) |
+| `monitor --demo [week\|drives] [--open]` | synthetic data: the tutor's week, paced, in under a minute; `drives`: a short goal-signal scenario |
 
 ## Testing a character
 - `lint`: contradicting habits; with `--props`, every rule proved or fuzzed;

docs/agents.md

@@ -216,5 +216,5 @@ Practical notes for agent use:
   (on Linux and macOS) `--priority low` keep it from crowding the agent's own process.
 - `--top`, `demo --live` and the hero screen draw only on a terminal; an agent's pipes never see them.
 - A persona's character rules in CI: `probbit persona lint PERSONA --props rules.json` proves each rule or, when the bound
-  cannot decide it, fuzzes it (exit 1 when one breaks); `probbit persona fuzz` / `prove` with `--json` give the documents
+  cannot decide it, fuzzes it (exit 1 when one breaks; exit 3 when one stays unproven: unknown and no counterexample found); `probbit persona fuzz` / `prove` with `--json` give the documents
   (docs/persona.md §5.6). They test the stance a host gets, not the words a model writes.