Dodaios is an open-source agent OS for your desktop: it runs your coding agents, browser and apps by voice or chat, on your own machine. Formerly Personal Jarvis.
Python agent-orchestrationai-agentai-coding-assistantclaude-codecoding-agentcomputer-usedesktop-applocal-firstmcppersonal-jarvisvoice-assistantvoice-coding
feat(verse): open every agent look from level 1
rubenluetke10-beep5fc0a5f38ea8
diff --git a/docs/agent-society/level-system.md b/docs/agent-society/level-system.mdindex b3cc64f662..21f839b081 100644--- a/docs/agent-society/level-system.md+++ b/docs/agent-society/level-system.md@@ -106,15 +106,9 @@ best unlocks.Agents also unlock the wearable looks of their profile symbol(`LOOK_UNLOCKS` in `rules.py`, ids from `companion/accessories.json`):-| Agent level | Looks |-|---|---|-| 1 | sunglasses, cap, headphones, coder hoodie |-| 2 | lab coat |-| 3 | cigar |-| 4 | suit and tie |-| 6 | top hat |-| 8 | tuxedo |-| 10 | crown |+All ten looks are open from level 1 (the maintainer's choice); raising a+value in `LOOK_UNLOCKS` turns that look back into a level-up reward, and+every surface below follows on its own.The snapshot carries them as `looks` and a level-up's `ProgressionAwarded`names them in `unlocked_looks`. The agent's level-up toast shows the agentdiff --git a/jarvis/progression/rules.py b/jarvis/progression/rules.pyindex 7aace7e77b..1a5e596ed4 100644--- a/jarvis/progression/rules.py+++ b/jarvis/progression/rules.py@@ -232,20 +232,20 @@ def rewards_between(kind: SubjectKind, before: int, after: int) -> list[Reward]:# ------------------------------------------------------------------ agent looks#: The agent level at which each wearable look unlocks-#: (``companion/accessories.json`` ids). Four are open from the start, so-#: every agent can be dressed at once; the rest arrive as level-up rewards.-#: A look an agent already wears is never taken away, whatever its level.+#: (``companion/accessories.json`` ids). The maintainer opened every look+#: from the start (2026-10-05); raise a value to turn a look back into a+#: level-up reward. A look an agent already wears is never taken away.LOOK_UNLOCKS: Final[dict[str, int]] = {"sunglasses": 1,"cap": 1,"headphones": 1,"hoodie": 1,- "lab_coat": 2,- "cigar": 3,- "suit": 4,- "top_hat": 6,- "tuxedo": 8,- "crown": 10,+ "lab_coat": 1,+ "cigar": 1,+ "suit": 1,+ "top_hat": 1,+ "tuxedo": 1,+ "crown": 1,}diff --git a/tests/unit/progression/test_rules.py b/tests/unit/progression/test_rules.pyindex ac63cf95f2..72f8cba8e4 100644--- a/tests/unit/progression/test_rules.py+++ b/tests/unit/progression/test_rules.py@@ -129,9 +129,6 @@ def test_subject_ids_name_their_kind():assert subject_kind("robot:x") is None-def test_agent_looks_start_open_and_climb_as_rewards():- assert sum(1 for level in LOOK_UNLOCKS.values() if level == 1) >= 3- assert max(LOOK_UNLOCKS.values()) <= 10- assert looks_between(1, 1) == []- assert looks_between(3, 4) == ["suit"]- assert set(looks_between(1, 50)) == {k for k, v in LOOK_UNLOCKS.items() if v > 1}+def test_every_agent_look_is_open_from_the_start():+ assert LOOK_UNLOCKS and all(level == 1 for level in LOOK_UNLOCKS.values())+ assert looks_between(1, 50) == []