98e60eab5b7cedc5ad0f58055dba5f93c1033974
19
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 
|
c910457817 |
v0.4.0 vNext — closed-loop world model + settlement contract (#29)
* feat(v0.4.0): vNext — closed-loop world model + settlement contract
Implements the vNext architecture from the research doc: demote the noisy
multi-rail planner in favour of a closed loop (world truth → invariant check)
plus a single utility-driven goal authority.
L1 services (fix no_drop / silent pathfinder hang first):
- InventoryLedger: diff-based "did I actually get it" verifier; acquire-food
now confirms via ledger.gainedSince instead of the unreliable count/event.
- MotionService.gotoSafe: wall-clock timeout + progress watchdog +
path_update(noPath/timeout) → structured {reached|stuck|timeout|nopath}.
L3 plan — unify the three competing rails (curriculum/manifesto/storyline):
- Settlement Contract: ordered M0–M9 milestones, each invariant-checked
against an authoritative world view (early steps delegate to the proven
curriculum; late game adds farming).
- InvariantChecker + predicate library; GoalManager selects the lowest unmet
milestone via utility argmax (food-urgency preempts, DEPS-style).
- Wired into the scheduler: bot.js precomputes snapshot.contract; reflex.js
consumes it in place of the storyline rail. Manifesto L0 still preempts.
Eval + robustness:
- Village Score (single 0..1 metric) on the snapshot + TUI "build" line.
- survive.dig-in skill + dusk_dig_in mode (exposed at night, no bed → cover).
- approach_block helper (GoalNear + lookAt, avoids GoalLookAtBlock #341).
+28 new tests (450 total green). LLM remains entirely off the tick path.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(v0.4.0): finish vNext plan — anti-loop, skill-graph, worldDelta diff, flee→motion
Completes the remaining v0.4.0 plan items and one fix motivated by a live
in-game observation (flee hanging 30s against a persistent zombie).
- flee → MotionService.gotoSafe: structured {stuck|timeout|nopath} in ~4s with
a blind-retreat fallback, instead of the observed 30s pathfinder hang + 3
watchdog replans. Movements setup guarded so it is unit-testable.
- QW5 anti-loop (runtime/anti-loop.js): same skill failing >=3x in 5min →
30min blacklist (reflex shouldSkip) + one-shot improvement_request
(bot.js drainFired -> writeProposal).
- 4.1 closed-loop worldDelta: runSkill snapshots inventory before execute and
attaches the real delta (_invObserved) to every successful result; opt-in
skill.expectGain asserts the claimed gain or returns world_unchanged.
- 3.6 skill-graph (Plan4MC): declarative requires/produces for ~20 skills;
prerequisitesMet/canRun/runnableFrontier; GoalManager annotates suggestions
with blockedBy when prereqs are unmet.
+22 tests (472 total green). Live smoke confirmed dig-in works and no new
errors; flee loop is what this commit's flee migration addresses.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
---------
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
|
15b6c11002 |
v0.3.1: survival behaviour overhaul — storyline, biome-aware scout, wedge-relocate, food/perf fixes, monitor TUI (#28)
* docs(v0.3.1): PRD — LLM prompt cost optimization
Design-only commit; no runtime changes. Spec for the next patch iteration.
Goal: cut per-advise() input tokens from ~800 to ≤300, preserving the
LLM's ability to produce valid registered skill ids and useful rationale.
Five proposed changes ranked by impact:
P1 Compact registry format (saves ~350t/call) — group by namespace,
comma-list ids, drop human titles. Default mode for advisor;
verbose mode kept for postmortem/reflect.
P2 Need-scoped registry (~50t additional) — show LLM only skills
relevant to the active Maslow need + always-available safety
skills (survive.flee, pillar-up, recovery.tunnel-out, explore.*).
P3 Snapshot pruning (~50t) — drop weather/experience/dimension/biome/
players from the user prompt; the LLM doesn't consult them.
P4 Prompt caching probe — check if TimeWeb passes through
prompt_tokens_details.cached_tokens. If yes, restructure prefix
to maximize cache hits (cached input is ~10x cheaper at OpenAI).
P5 Per-trigger cost telemetry in scripts/list-improvements.js --stats:
avg_in / avg_out / cost_₽ / share% per trigger_reason, using
TIMEWEB_PRICE_IN_RUB_PER_M and TIMEWEB_PRICE_OUT_RUB_PER_M env.
Trigger: TimeWeb admin panel after first day of v0.3.0 live showed
34K tokens / day at low activity. At cap budget that projects to
~480₽/month (101₽/M in, 608₽/M out for gpt-5.4-mini). Manageable
but the savings are mostly free — repeated infra tokens, not signal.
All changes are additive; runtime behaviour stays the same. If the
LLM produces worse advice with the compact registry, flip back via a
single constant in fast-advisor.js.
Acceptance: re-run scripts/check-timeweb.js probe 3 — expect
tokens_in ≤ 300 (was ~800). Live for 1h, check --stats: avg_in ≤ 300
per trigger group. Existing 360 tests still green.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(v0.3.1): storyline — canonical Minecraft survival quest
The bot has been stuck in a loop for two days:
acquire-food (fail: no nearby food) → explore.far → pillar-up (fail) → repeat
Diagnosis: manifesto + LLM advisor both correctly identify "you need
food" but neither expresses *what concretely to do next*. Manifesto is
a priority ladder (need-detection), not a narrative arc.
This commit adds the missing narrative layer — an ordered list of
operational steps that mirror the vanilla Minecraft survival path:
1. orient_self — Понять где я
2. first_wood — Собрать 8 поленьев
3. crafting_basics — Сделать верстак и палки
4. first_tools — Деревянные орудия
5. first_food — Найти первую еду
6. shelter_minimal — Простой шелтер с кроватью
7. stone_tier — Каменные орудия
8. food_security — Запас еды на 16+
9. iron_age — Железо и печь
10. settle_base — Постоянная база
11. village_grow — Развивать деревню (ongoing)
Each step has:
- completed(snapshot) → bool — detects achievement from snapshot
- suggestSkill(snapshot) → { skillId, args? } — concrete next dispatch
- emergencyPause(snapshot) → bool — defers to manifesto L0 alive
emergencies (low HP near hostile, lava under foot, food = 0)
- narration_ru — chat-friendly Russian one-liner spoken on entry
Components:
- runtime/goal/storyline.js — 11-step canonical quest catalogue
- runtime/goal/state.js — pickCurrentStep(snapshot) walks the list,
returns first non-completed step + its suggestion. 3s cache.
Validates suggestSkill's skillId against the live registry.
- runtime/reflex.js — curriculumReflex dispatch priority is now:
1. manifesto (L0 alive emergencies always win)
2. storyline (concrete operational subgoal)
3. curriculum plan (legacy fallback)
Tests pass ctx.disableStoryline=true for isolation.
- runtime/bot.js — snapshot.storyStep populated each tick so
chatter/advisor/reflect observers see the same view.
- runtime/coach/fast-advisor.js — buildUserPrompt now embeds the
current step + its suggested skill, so LLM advice is anchored
("step 5 first_food, storyline wants survive.acquire-food, but
recent dispatches show it's failing — try explore.far + scout").
- runtime/coach/advisor-trigger.js — forwards ctx.storyStep into
advise() and logs step id at trigger time.
- runtime/coach/reflect.js — reflection prompt includes storyline
progress so 30-min self-assessment is anchored.
- runtime/persona/chatter.js — narrates step.narration_ru on
transition. Rate-limited via existing maybeNarrateRaw().
New operator CLI:
- scripts/show-story.js — fetches the live snapshot via IPC sock and
prints step progress with ✓/→/ markers, current skill, inventory.
Falls back to --plain catalogue view when bot offline.
Token cost impact: ~+30 input tokens per advise() call (one extra
line in user prompt). Trivial vs the value of grounding LLM advice
in a concrete narrative.
Operator usage:
node scripts/show-story.js # live progress + which step + why
node scripts/show-story.js --plain # static catalogue of all 11 steps
Tests: 376 green (was 360, +16 storyline tests).
Also in this branch (already committed): dev/v0.3.1/PRD.md —
LLM prompt cost optimization design doc.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* fix(v0.3.1): storyline beats manifesto L1+ (only L0 alive emergencies override)
Found in live logs after the previous commit deployed:
storyline: step 1/11: orient_self → explore.wander
advisor-trigger: firing because wedged (planned=survive.acquire-food, ...)
Manifesto was still picking survive.acquire-food (L1 food) over the
storyline's orient_self → explore.wander. That's the wrong precedence —
storyline expresses a *concrete operational subgoal* and L1+ manifesto
needs are just "you'd benefit from food" priorities, not emergencies.
New dispatch precedence in curriculumReflex:
1. manifesto L0 (alive emergencies: lava, low-HP+hostile, food=0)
2. storyline (concrete narrative subgoal — beats L1+ manifesto)
3. manifesto L1+ (fallback when storyline has no concrete suggestion)
4. curriculum plan (legacy fallback)
This way the bot starts following the narrative arc even while
manifesto's L1 food is technically unsatisfied — orient_self runs to
completion before pursuing food explicitly. Storyline already handles
food as step 5 (first_food), so we're not skipping it.
Tests: 378 green (+2 priority-ordering tests):
- L0 manifesto emergency: upstream reflex (defend/modes) catches before
curriculum dispatch
- storyline beats manifesto when both have suggestions: well-fed bot
with logs → craft.planks (storyline crafting_basics), not gather.logs
(manifesto L2)
- updated "manifesto fallback" test to require disableStoryline=true
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(tui): fullscreen monitor-only TUI (opencode-style)
Replaces the old tui/tui.tsx hotkey-heavy dashboard with a read-only
observability screen. Operator actions live in scripts/* now —
TUI is for watching, not driving.
Layout (top to bottom, all auto-resizing to terminal):
1. Header — MC/IPC status, pos, HP, food, day/night, hostiles
2. Storyline — current step + 11-step quest map (✓/→/○)
3. Activity — last N skill dispatches (colour by outcome)
4. MC Chat — last N chat lines (cyan for bot, yellow for players)
5. Advisor — last N LLM recommendations (trigger + outcome + tokens)
6. Improvements — open requests from knowledge.improvement_requests
7. Footer — 24h token usage + cost in ₽ + q-to-quit
Data sources:
- IPC sock: snapshot frames, log frames, chat frames (push)
- SQLite knowledge.db: advisor_recommendations + improvement_requests
polled every 5s (pull)
Token cost displayed live using TIMEWEB_PRICE_IN_RUB_PER_M /
TIMEWEB_PRICE_OUT_RUB_PER_M env vars (defaults: 101 / 608 for
gpt-5.4-mini).
Switches:
- npm run tui → new monitor (this file)
- npm run tui:legacy → old action-driven tui/tui.tsx (kept for now)
Implementation notes:
- Uses ink + alternate-screen-buffer ANSI for proper "opencode-feel"
fullscreen behaviour; restores prior terminal contents on quit.
- Skips alt-screen and useInput when stdin/stdout isn't a TTY
(smoke tests, piped output) — both gracefully degrade.
- Stable React keys via per-event uid counter, avoids reconciler
duplicate-key warnings as logs/chat/dispatches stream in.
- Resize handled via 1s stdout-dimension poll, NOT direct
'resize' listener (which conflicts with ink's own listener and
triggers MaxListenersExceededWarning).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* ui(tui): compact 4-section monitor (was 6) — fits 1080p without zoom
Operator reported the TUI overflowed the screen unless terminal was
zoomed way out. The 11-step storyline list alone was eating ~13
rows, and each advisor/improvement entry took 2-3 rows. Now:
- Header + storyline collapsed into one panel (2 lines):
line 1: pepa · ●MC ●IPC · 1m50s · pepa_bot · (697,61,702) · HP 20 · food 5 · ☀ · ⚔60(creeper@58b)
line 2: story ▓▒░░░░░░░░░ 1/11 orient_self · Понять где я → explore.wander
The 11-step ladder is now a unicode progress bar (▓ done, ▒ current,
░ pending) — same info, fits in one row.
- Advisor entries: one line each instead of two.
✓ wedged_60s → survive.flee 802t 1900ms
(outcome mark / trigger / target skill / tokens / latency)
- Improvements entries: one line each instead of two.
#1 P2 ×3 Add craft.iron-pickaxe skill
Description dropped from the row — use `node scripts/list-improvements.js`
for full text.
- Sections: 4 (was 6).
[header+story] · [activity | chat] · [advisor | improvements] · [footer]
Tested on a typical 1080p terminal — fits comfortably without zoom.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(v0.3.1): real survival patterns — biome-aware scout, wedge-relocate, escape-pit-safe
Operator reported the bot wandered the same 50×50 patch for 2 hours
without making any progress toward food. Diagnosis showed three root
causes; this commit addresses all five open improvement_requests
the LLM (postmortem + tuner) flagged automatically.
Research basis (`Voyager`, `Plan4MC`, `GITM`, `Mindcraft`):
- Coverage / commit-to-cardinal exploration when local scan fails
- Biome-aware strategy switching using a static affordance table
- Wedge detector above the skill layer that triggers RELOCATE not
RETRY (per-skill stuck checks reset on re-entry — useless)
- Time-in-region bbox heuristic + need-duration AND skill-cycle gate
Concrete changes:
1. `runtime/goal/storyline.js`
- orient_self.completed: added timeout fallback (HP=full + session
>120s → done) so barren biomes don't block the bot on step 1.
Closes improvement #2 'Нет навыка оценки когда сменить район'.
- first_food.suggestSkill: now picks survive.scout-food (new) when
no passive mob is nearby; falls back to survive.acquire-food only
when something is in immediate range.
2. `runtime/biome-affordances.js` (new)
- Static table: 40+ biomes → {has_passive_mobs, has_trees,
has_water, has_crops, livable}.
- Unknown biomes return optimistic defaults to avoid regressions.
- Closes improvement #1 'Нет навыка целевого поиска еды по биому'.
3. `runtime/skills/scout-food.js` (new — survive.scout-food)
- Tiered strategy: biome check → scan 32 → scan 64 → commit a
cardinal for 200 blocks rescanning every 16. On cardinal
exhaustion, returns code:"exhausted" so the curriculum can
escalate to village.relocate.
- In barren biomes (desert/ocean/snowy_plains) the scan is
SKIPPED — bot walks straight toward the nearest neighbour
biome that affords passive mobs (8-direction biome probe at
radius 64).
4. `runtime/awareness/wedge-detector.js` (new)
- Rolling 10-min position bbox tracker. observe() called every
tick; isWedged() returns true when bbox<50 AND active need
unmet >5min AND skill cycles ≥3.
- markRelocationStarted() suppresses further wedge firings
until the bot has displaced ≥200b — prevents stack overflow
of relocate calls.
- Lives ABOVE the skill layer (in runtime/reflex.js), because
any per-skill stuck check resets on re-entry.
5. `runtime/skills/relocate.js` (new — village.relocate)
- 300-block walk in least-recently-used cardinal (per-incident
memory in ctx.recentRelocations).
- Re-paths every 32 blocks, soft-tolerates pathfinder failures
(3 consecutive throws → exit with code:"stuck_in_place").
- Closes improvement #2 + #4 ('low success rate trigger').
6. `runtime/skills/escape-pit-safe.js` (new — recovery.escape-pit-safe)
- Surveys 4 cardinals AND ceiling height before committing.
Picks the direction with most open blocks (≥3, no lava).
Falls through to pillar-up only if ceiling clear ≥4b. Returns
code:"no_strategy" if both blocked so curriculum can escalate
to relocate.
- Closes improvement #3 'Нет навыка для безопасного выхода'.
7. `runtime/reflex.js`
- Wedge detector wired before manifesto/storyline. If wedge.wedged
is true, dispatches village.relocate directly and returns —
bypasses every other branch.
- ctx.disableWedge flag for tests.
8. `runtime/coach/advisor-trigger.js`
- LLM provider outage backoff: 3 consecutive http_400 / timeout /
network_error → suppress advisor for 10 min. Today's TimeWeb
gpt-5.4-mini was 400'ing for an hour straight; we were spending
trigger budget on dead calls. Closes improvement implicit gap
in #5.
9. `runtime/bot.js`
- Tracks botSpawnedAt; snapshot._sessionMs exposed for storyline
orient_self timeout fallback.
Tests: 396 green (was 378, +18):
- runtime/biome-affordances.test.js — 8 tests
- runtime/awareness/wedge-detector.test.js — 9 tests
- runtime/goal/storyline.test.js — 1 new test (orient_self timeout)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* fix(coach/trigger-tuner): crash after ~1h — runOnce is sync, not a Promise
The live bot died overnight with:
TypeError: runOnce(...).catch is not a function
at trigger-tuner.js:42 → [supervisor] child exited code=1
attach() wrapped the timer body as `runOnce().catch(...)` but
runOnce() returns a plain {ok, flagged, ...} object (pure SQL, no
await). The first tuner tick (60min after spawn) threw → killed the
whole bot process. Never surfaced before because the bot rarely ran
uninterrupted for a full hour during development.
Fix: guard the synchronous call with try/catch, matching how
persona/chatter.js already does its sync tick. (postmortem.drainOnce
and reflect.runOnce ARE async, so their .catch is correct — audited.)
Regression test added: captures the setInterval callback and invokes
it synchronously, asserting it does not throw.
Tests: 397 green.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* fix(v0.3.1): mechanical food/stuck fixes — bot reaches the chicken now
The wedge wasn't only in the manifesto layer; several mechanical bugs
kept the bot in a dead random-walk:
- storyline / manifesto / curriculum: "local food" now means an edible
passive mob within <=32 blocks. A distant chicken or a cod no longer
fools the bot into dispatching acquire-food (which then fails on
no_path). Long-range food goes through scout-food instead.
- scout-food: partial approach to a target now counts as progress
(approached_target, e.g. moved:14); a blocked heading is NOT counted
as movement; added blind/tunnel fallback so it doesn't die when the
pathfinder can't route cleanly.
- acquire-food: on no_path it now also tries a blind/tunnel approach to
the animal; no_drop routes back into food scouting instead of giving
up.
- explore.far / relocate / flee: fewer false "done" results (micro-steps
no longer counted as success), more genuine escapes from stuck.
- scripts/show-story.js: live IPC now actually renders the current
storyline step.
Verification: scripts/lint-patch.js clean; npm test 404/404 green; bot
relaunched in tmux `pepa`. Live logs show real progress — bot switched
to survive.scout-food, approached the chicken (approached_target
moved:14), then reached survive.acquire-food: hunting chicken. Food
isn't fully closed yet but the remaining issue is concrete pickup/drop,
not dead random-walk.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* fix(v0.3.1): sated bot stops chasing food + perf-leak + fuzzy improvement dedup
Third day of "bot just walks back and forth burning tokens". Root
causes were mechanical, not the manifesto:
1. SATED BOT CHASING FOOD (the big one)
Bot had food=17 (nearly full) but storyline first_food + manifesto
L1 required 2+ food ITEMS in inventory, so it looped scout-food /
acquire-food for hours instead of working. Now both treat a hunger
bar >= 14 (SATED_FOOD) as satisfied even with empty food inventory —
a full bot chops wood / makes tools and grabs food opportunistically,
only hard-pursuing food when actually hungry (< 14).
manifesto/needs.js foodDetect + goal/storyline.js first_food.completed.
2. perf_hooks MEMORY LEAK (overnight OOM suspect)
"MaxPerformanceEntryBufferExceededWarning: 1,000,001 measure entries".
mineflayer/pathfinder emit perf marks we never consume. Added a
60s reaper in bot.js (performance.clearMeasures/clearMarks). unref'd.
3. IMPROVEMENT QUEUE SELF-DUPLICATING
The LLM re-filed closed gaps with reworded titles (#5/#8/#9 were
dupes of implemented #1/#2/#3). Exact-title dedup missed them.
Replaced with token-set fuzzy match (isDuplicateTitle): jaccard>=0.75
OR >=3 shared meaningful tokens with jaccard>=0.5. Also: a re-filed
gap that's already implemented/rejected is NOT resurrected as a new
open row. Cleared all 5 open requests (now genuinely implemented).
Also confirmed (no change needed):
- canDig=true is a DELIBERATE codebase-wide choice ("without it the bot
gets permanently stuck", actions.js). The stale memory recommending
canDig=false is updated. ViaBackwards dig works partially (dug:1
moved:1.8 observed); false would trap the bot in every pit.
- scout-food already has blind/tunnel fallback + 12s step timeout
(operator's earlier edits) so trapped-pathfinder degrades instead of
hanging 30s.
Tests: 407 green (was 404). Updated needs/state/storyline tests for the
SATED_FOOD threshold; added fuzzy-dedup + tokenize/jaccard tests.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
---------
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
|
7d96e44804 |
v0.3.0: Maslow + Awareness — self-learning bot with needs ladder, event-driven reflex, and TimeWeb fast advisor (#27)
* v0.3.0-rc.1: live skill registry + fast advisor scaffold
Roots out the v0.2.x failure mode: Pi-extracted lessons routinely named
hallucinated skill ids (relocate.surface, choose.safe.surface,
survive.shelter, gather.visible_log, …). All 47 Pi-lessons in the live DB
had applied_count=0 because normalisePreferSkill couldn't find them.
Fix:
1. runtime/skill-registry.js — single source of truth derived from
skills/index.js. Exports listSkillIds, isRegistered, and a
prompt-ready block (skillRegistryPrompt) grouped by namespace.
2. Pi prompts (coach/postmortem, coach/reflect) embed the live registry
with a "USE ONLY THESE, never invent" instruction. Lessons are
filtered at write-time too — anything not in the registry and not a
known mode name gets dropped.
3. coach/advice.js — normalisePreferSkill now returns null for unknown
ids, hardening consult() against any hallucinations that slip
through. Warn-logged for visibility.
Also lays the LLM substrate for the rest of v0.3.0:
- runtime/llm/provider.js — OpenAI-compatible chat client. Configured
via PEPA_FAST_LLM_{BASE_URL,API_KEY,MODEL,TIMEOUT_MS}. Safe no-op
unless API_KEY is set. Supports JSON-mode.
- runtime/coach/fast-advisor.js — tactical advisor tier (scaffold).
Exposes advise() that asks the fast LLM what to do RIGHT NOW when
the reflex is wedged/stuck. Rejects hallucinated skill ids using the
registry. Rate-limited 6/h, 30s cooldown. Not auto-triggered yet —
wired into reflex in rc.3 (awareness layer).
Tests: 279 green (+24 vs rc.3): 5 registry, 9 provider, 10 advisor.
See dev/v0.3.0/PLAN.md for the full iteration design (manifesto needs
ladder, event-driven awareness, skill pre-emption) and STATUS.md for
shipped/pending tracking.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* v0.3.0-rc.2: manifesto / needs ladder L0-L10
Adds an explicit hierarchical needs catalogue that the reflex consults
on every tick. The bot now pursues tangible intermediate goals (food,
wood tools, shelter, stone tools, ...) instead of inheriting whatever
the curriculum thought was "next".
Ladder:
L0 alive HP>5, food>0, not in lava, not panic-near hostile
L1 food ≥6 food items in inventory (or sated + any food)
L2 tools_wood wooden_pickaxe + wooden_axe + wooden_sword
L3 shelter_basic bed placed nearby or in inventory
L4 tools_stone stone-tier triplet
L5 armor_basic any chestplate (pursue=null until craft.leather-*
lands; ladder gracefully skips)
L6 food_security ≥16 food items
L7 tools_iron iron-tier triplet (pursue=gather.stone for now)
L8 armor_iron iron chestplate (pursue=null for now)
L9 village_seed bed + chest in nearby blocks
L10 village_full never detected, falls through to curriculum
Each need has detect(snapshot) → bool and pursue(snapshot) →
{skillId, args} | null. The ladder picks the LOWEST unsatisfied
pursuable need. Needs whose pursue is null get recorded as
blockedNeeds and the walk continues — no stalling on missing skills.
Wired into curriculumReflex: manifesto takes precedence over
curriculum.plan when it has a concrete suggestion. Tests can pass
ctx.disableManifesto=true to exercise the curriculum branch
in isolation (existing reflex tests keep passing this way).
Pi self-reflection prompt now includes
"activeNeed (Maslow ladder L0-L10): L2 tools_wood → gather.logs"
so Pi advises at the right level instead of giving generic guidance.
skillId returned by pursue() is validated against the live registry
(rc.1 plumbing) — manifesto cannot accidentally dispatch a
hallucinated skill name.
Tests: 315 green (was 279 on rc.1, +36 new):
- runtime/manifesto/needs.test.js — 24 tests (per-need detect/pursue,
helper sums)
- runtime/manifesto/state.test.js — 10 tests (ladder walk, hostile
takeover at L0, armor skipping, caching)
- runtime/reflex.test.js — 2 integration tests (manifesto overrides
curriculum plan; well-fed bot pursues tools_stone)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* v0.3.0-rc.3: event-driven awareness + skill pre-emption
Adds a reactive layer on top of the polling reflex. The bot now
notices environmental shocks (forced moves, HP plunges, hostile
spawns) within ~100ms instead of waiting for the next DISPATCH tick,
and the in-flight skill is preempted so the next reflex cycle can
re-plan against the current world state.
This is the rc that wires the "rc.1 plumbing + rc.2 manifesto" into
a feedback loop:
- awareness fires preempt → dispatch aborts
- reflex tick re-evaluates → manifesto walks the ladder
- new dispatch picks the right skill for the new world state
Pieces:
- runtime/awareness/events.js (new) — bot.on listeners:
- move: single-tick Δposition ≥ 5 blocks → forced_move flag + preempt
- health: HP drop ≥ 2 → health_plunge flag + preempt
- entitySpawn: hostile mob within 12 blocks → hostile_added + preempt
- blockUpdate: nearby block change → env_changed flag (no preempt,
throttled 800ms; otherwise gather skills would self-preempt
every dig)
- runtime/skills/index.js — RUNNER_CODES.PREEMPTED + raceWithAbort()
wraps every execute() against ctx.abortSignal. Existing skills get
preemption for free; they don't have to check the signal manually.
- runtime/bot.js:
- dispatchAction creates a fresh AbortController per dispatch and
stores it on reflexCtx.currentAbort
- attachAwareness fires controller.abort() when something disrupts
the active skill; runSkill returns code: "preempted" and the
reflex moves on
- reflexCtx.lastPreempt records the most recent shock
Tests: 332 green (was 315 on rc.2, +17 new):
- runtime/awareness/events.test.js — 12 tests (each event type +
thresholds + throttling + passive-mob filter)
- runtime/skills/contract.test.js — 3 abortSignal tests
(mid-flight, pre-armed, clean signal)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* chore(.env): add PEPA_FAST_LLM_* placeholders for v0.3.0 fast advisor
Empty values keep the fast-advisor tier disabled (safe no-op). Fill
in BASE_URL + API_KEY + MODEL to enable. TimeWeb-style endpoint
example included.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* chore(v0.3.0): rename fast-LLM env vars to TIMEWEB_* (match other projects)
Aligns with the user's other repos (proso) which use TIMEWEB_API_GROK /
TIMEWEB_URL_GROK. Single naming convention across projects avoids the
'which env var was it for this repo' mental tax.
PEPA_FAST_LLM_BASE_URL → TIMEWEB_BASE_URL
PEPA_FAST_LLM_API_KEY → TIMEWEB_API_KEY
PEPA_FAST_LLM_MODEL → TIMEWEB_MODEL
PEPA_FAST_LLM_TIMEOUT_MS → TIMEWEB_TIMEOUT_MS
Provider still works with any OpenAI-compatible endpoint — TimeWeb is
the default but the variable name doesn't lock us in. Tests + docs +
.env / .env.example updated.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(scripts): TimeWeb smoke test + bump default LLM timeout to 20s
scripts/check-timeweb.js — three probes: plain text, JSON mode, full
fast-advisor stack (registry injection + skill validation). Loads .env,
prints {ok, latency, reply preview} for each. Doesn't touch bot state.
Bumped DEFAULT_TIMEOUT_MS 8s → 20s in runtime/llm/provider.js. TimeWeb's
hosted agent endpoint takes 5-15s for the fast-advisor prompt
(registry block + snapshot context), so 8s was producing spurious
timeouts. OpenAI direct returns much faster; env var TIMEWEB_TIMEOUT_MS
overrides if needed.
Smoke verified live (PR #27 branch):
probe 1: 6.3s, plain prompt → "pepa hears you"
probe 2: 5.4s, JSON mode → {"alive":true,"name":"pepa"}
probe 3: 14.9s, advise() → action=switch_skill, skill=recovery.tunnel-out
(correct registered skill, sensible rationale — registry
injection successfully prevents hallucination)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(v0.3.0): auto-trigger fast-advisor + token usage tracking
Closes the awareness → LLM → action loop that the rc.1/2/3 sequence
left as a followup. When the bot is wedged, looping, or just suffered
a preempt-then-retry, the reflex fires advise() in the background;
when the recommendation lands it overrides the next dispatch.
Async by design: advise() takes 5-15s on TimeWeb's hosted endpoint —
too slow for a synchronous reflex tick. tickAdvisor() is fire-and-
forget, the result lands on ctx.advisorRecommendation, and the *next*
tick reads and consumes it. Recommendations age out after 60s.
Components:
- runtime/coach/advisor-trigger.js — policy + async fire path
- tickAdvisor(ctx, {plannedSkillId}) checks three triggers:
1. wedged > 60s (no significant move)
2. last 4+ dispatches are the same skill AND it's planned again
3. preempt within last 30s + same skill being retried
- 90s trigger cooldown, single-in-flight guard
- consumeFreshRecommendation(ctx) reads/clears the cache
- runtime/reflex.js — curriculumReflex calls tickAdvisor() every tick
and consumes a fresh recommendation BEFORE dispatching. ctx flag
disableAdvisor=true for tests.
- runtime/bot.js — dispatchAction maintains a rolling 8-slot
reflexCtx.recentSkillIds for the loop-detection trigger.
Token usage:
- runtime/llm/provider.js — normaliseUsage() reads OpenAI/TimeWeb-
style {prompt_tokens, completion_tokens, total_tokens} from the
response. Returned on every complete() result and logged at info
level as "in=Nt/out=Mt".
- runtime/coach/fast-advisor.js — getUsageSnapshot() aggregates
total tokens across all calls in the session.
Measured on live TimeWeb endpoint (gpt-5.4-mini agent):
per call: ~705 input + 45 output = ~750 tokens
rate limit: 6 calls/hour
worst case at full budget: ~108K tokens/day
estimated cost (OpenAI gpt-5-mini reference price): ~$0.60/month
Well within any reasonable budget — model can run hot 24/7.
Smoke verified: scripts/check-timeweb.js probe 4 produces
trigger fired: true (wedged_90s)
recommendation: recovery.tunnel-out
rationale: "Stuck wedged for 90s; exploration is failing."
latency: 5302ms
Tests: 345 green (was 332, +13 advisor-trigger).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(v0.3.0): paradigm shift — TimeWeb-only LLM + persistent advisor trail + improvement queue
This is the rc.4 batch the user requested:
1. Emergency triggers (low HP + close hostile, lava-under-foot)
bypass the long cooldown so the LLM is consulted BEFORE the bot
dies, not after.
2. Active manifesto need is now included in the advisor user prompt
— the LLM picks suggestions that satisfy the bot's current
concrete need (L2 tools_wood → "gather logs nearby" not
"explore further").
3. Every advisor recommendation is persisted to SQLite
(advisor_recommendations table) with full token usage. The
reflex marks 'applied=1' when it dispatches and updates
outcome_ok/code when the dispatch completes. Ground truth for
"is the LLM actually helping" lives in the DB, not in logs.
4. Pi CLI is OUT of every background loop. coach/postmortem and
coach/reflect now go through the same TimeWeb endpoint
fast-advisor uses, via the shared coach/llm-call.js helper.
Pi is reserved for manual operator commands.
5. The LLM (postmortem, reflect, advisor) can flag "structural
gaps" — missing skills/features the operator should implement.
These land in the new improvement_requests table. Dedup by
title bumps `votes` instead of inserting duplicates so the
queue doesn't bloat. Operator views via
`node scripts/list-improvements.js`.
6. A deterministic trigger-tuner runs hourly: reads 24h of
recommendation stats, flags triggers whose success rate is
below 25% (sample ≥ 5) or whose prompts are expensive (>1000
input tokens) with mediocre payoff. Improvements get
source="tuner", category="tuning". No LLM call.
New files:
runtime/coach/llm-call.js — askAnalytical() helper
runtime/coach/trigger-tuner.js — stats → improvements
runtime/coach/trigger-tuner.test.js
scripts/list-improvements.js — operator CLI
Schema additions:
advisor_recommendations: id, ts, trigger_reason, planned_skill,
recommended_skill, action, rationale, active_need, tokens_in,
tokens_out, latency_ms, applied, outcome_ok, outcome_code, outcome_at
improvement_requests: id, ts, source, category, title, description,
context, priority, status, duplicate_of, votes, implemented_at, notes
Renamed env-var consumers:
Pi-coach drainOnce({ askPi }) → drainOnce({ askAnalyticalFn? })
Pi-reflect runOnce({ askPi }) → runOnce({ askAnalyticalFn? })
bot.js attachCoach/attachReflect no longer pass askPi
attachTuner() added to bot.js spawn handler
lessons.source 'pi-coach' → 'timeweb-coach'
lessons.source 'pi-reflect' → 'timeweb-reflect'
Token cost measured live:
~705 input + 45 output = ~750 total per advisor call
worst case @ 6 calls/hour rate cap = ~108K tokens/day
OpenAI gpt-5-mini reference price: ~$0.60/month
Operator usage:
node scripts/list-improvements.js # open queue
node scripts/list-improvements.js --stats # advisor performance
node scripts/list-improvements.js --done 17 "shipped in 0.3.1"
node scripts/list-improvements.js --reject 18 "duplicate"
Tests: 360 green (was 332, +28 new).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
---------
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
|
865aae1213 |
v0.2.0-rc.3: pillar-up escape + advice in fallback + danger POI (#22)
* v0.2.0-rc.3: pillar-up escape + advice everywhere + danger POI
Closes the gap rc.2 left open. Live observation showed:
- Pi-coach extracted 5 high-quality lessons (do not explore.far at
night near zombies, etc.) but none of them fired (applied_count=0
across the board). Root cause: dispatcher consulted advice only on
the main curriculum path; the bot was falling into the wander/
explore.far FALLBACK after each gather attempt bailed, which
bypassed consult().
- Bot was wedged in a pit on (608, 90) with stone walls. recovery.
tunnel-out kept failing ("Digging aborted") because mining stone
with fists takes ~10s/block; pathfinder watchdog kills it.
This patch:
1. survive.pillar-up (runtime/skills/pillar-up.js) — new escape skill.
Places a placeable block under the bot and jumps onto it; repeats
up to 8 steps. No pickaxe required. Works in dirt/cobble/planks/
sand/gravel/wool/etc. The bot's vertical exit from any pit it can
stand in.
2. Wedged-emergency reflex (runtime/reflex.js). At the top of
curriculumReflex, if noProgressReason is wedged-like AND position
hasn't shifted ≥16 blocks in 60s AND no hostile in 6m AND pillar
block in inventory → dispatch survive.pillar-up. 2-min cooldown
between attempts.
3. consult() now also runs on the WANDER/explore.far fallback path
(runtime/reflex.js curriculumReflex). Pi-coach lessons can finally
take effect. If the fallback skill is overridden to a non-eligible
skill but the bot has a placeable block, falls back to pillar-up.
Outcomes feed reportAdviceOutcome so confidence stays grounded.
4. recordPOI("danger") on death (runtime/coach/postmortem.js). Spatial
memory now flags where the bot died, expires after 6h. POI table
was empty in rc.2.
5. SAFE_OVERRIDES extended (runtime/coach/advice.js): adds
survive.pillar-up and village.choose-base so coach lessons can
route there.
Tests: 255/255 green (+9 pillar-up).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* v0.2.0-rc.3 fixup: relax wedged-escape trigger
Drop the WEDGED_REASONS check — noProgressReason is a string that
may or may not be set when the bot is stuck. Fire pillar-up purely on
"no horizontal progress ≥ 60s, no hostile in 6m, placeable block in
inv". Pillar-up is a constructive no-op when it's not needed (places
one dirt under self) so the false-positive cost is small.
Live observation: rc.3 was deployed and bot was wedged with tunnel-out
repeatedly aborted on stone, but wedged-escape never fired because
the runtime's noProgressReason wasn't in my whitelist. Removing the
gate lets the trigger actually engage.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
---------
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
|
e84148d189 |
v0.2.0-rc.2: P0 hardening — Pi headless, test state isolation, advice fixes (#21)
P0 (correctness):
1. PEPA_HEADLESS=1 guard in extensions/mineflayer-bridge.ts. When `pi -p`
spawns a subprocess (banter, coach, planner, reflect, auto-patch), the
bridge no longer attempts a second MC connect — the hybrid runtime
already owns the nickname. runtime/pi-bridge.js sets the env var on
every spawn. Root cause of the "two pepa_bot's racing for the slot"
bug seen in reply-pi stderr.
2. Test state isolation in runtime/config.js. When running under the node
test runner (detected via execArgv/argv) — or when PEPA_STATE_DIR is
set — stateDir redirects to /tmp/pepa-test-state-<pid>/. log.js,
scenario-memory, world-journal, and knowledge.db all follow.
`npm test` no longer pollutes live scenarios.jsonl, world-journal.jsonl,
or daily log files. Verified empirically: post-fix run added 0 test
rows to the live scenarios file. Cleaned ~550 historical test rows
from live state in the same change.
3. defendReflex outcome reporting (runtime/reflex.js). Previously a
creeper-rule override marked the lesson succeeded=false BEFORE the
flee skill returned. Now dispatchDefendFlee accepts {lessonId} and
the onComplete fires reportAdviceOutcome with the actual flee result.
4. Mode-name → skill-id translation in runtime/coach/advice.js. Pi-coach
occasionally returns prefer_skill values that are mode names
("night_shelter", "self_preservation", "hunger"). normalisePreferSkill
maps these to SAFE_OVERRIDES entries before dispatch. Also handles
"tunnel-out", "survive_flee", "survive flee" shapes.
New behavior:
5. Self-reflection loop (runtime/coach/reflect.js). Every 30 min, the
bot asks Pi: "Are you making progress, or stuck in a loop? What
should you do differently?" Pi answers with a verdict
(progress/loop/recovering/idle/emergency), summary, next-action, and
0-N new lessons. The reflection is written to
state/<host>/reflections/<ts>.md and lessons land in the DB with
source="pi-reflect". Rate-limited to 2 calls/hour. Wired through
bot.js with the existing askPi + lastSnapshot accessor.
Tests: 246/246 green (+9 new: 6 advice mode-name + 4 reflect).
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
mayatnikovandClaude Opus 4.7
|
e183aaec4e |
feat(runtime/coach,reflex): retrieval-augmented dispatch via learned lessons
This closes the learning loop. Lessons in knowledge.db now actually
influence behaviour:
- runtime/coach/advice.js: consult({plannedSkillId, snapshot}) reads
knowledge.topAdvice() and returns 'override' / 'avoid' / 'proceed'.
When a lesson says "avoid <skill>" with prefer="survive.flee" (etc.),
the dispatcher swaps in the alternative.
- runtime/reflex.js:
* curriculumReflex now consults advice before dispatch; on 'avoid'
backs off the planned skill + sets wander hint; on 'override'
dispatches the lesson's preferred alternative.
* defendReflex (dist≤4 melee branch) consults advice too — so a
creeper at 4m honours the starter rule "attack creeper → flee".
Failure outcomes feed back via markApplied so confidence stays
grounded.
SAFE_OVERRIDES whitelist contains only known runSkill targets
(survive.flee, survive.sleep, survive.eat, recovery.tunnel-out,
explore.far/wander, village.build-shelter); unknown prefers fall back
to plain 'avoid'.
7 advice tests; total suite 237 green.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
mayatnikovandClaude Opus 4.7
|
86e5294bb8 |
chore: snapshot pre-v0.2.0 WIP (pathfinder/reflex/metrics/skills improvements)
Baseline for the v0.2.0 self-learning iteration. All 205 tests pass on this state. Subsequent commits in this branch layer the knowledge base, post-mortem coach, and persona narration on top. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> |
||
|
mayatnikov
|
f13799de28 | fix(runtime/reflex): retreat after repeated melee clears | ||
|
mayatnikov
|
b23aad2128 | fix(runtime/reflex): verify melee clears hostile | ||
|
mayatnikovandClaude Opus 4.7
|
4ae63dabe1 |
feat(runtime): v0.1.0 — adopt Voyager critic + Mindcraft modes/library/lint
Five concrete patterns from Voyager and Mindcraft, applied in our shape
without abandoning the git-as-evolution-substrate that makes pepa
distinct. Plus a first multi-agent surface so two bots from the same
repo can share intent.
1. runtime/critic.js (Voyager critic.txt)
- Spawns `pi -p` with a JSON-only critic prompt before a proposal is
written. {reasoning, success, critique}.
- success=true short-circuits the proposal (bot recovered between
detector tripping and now), saving Pi tokens on false positives.
- critique is spliced into the proposal body via attachCritique() so
the downstream auto-patcher has a sharp spec.
- Graceful: pi missing / timeout / unparseable JSON → proposal still
filed without the critic block.
2. scripts/lint-patch.js (Mindcraft coder._lintCode)
- Pre-flight gate between Pi commit and npm test: node --check, dynamic
import (catches missing named exports), regex extraction of
runSkill("id") calls cross-checked against the live registry.
- Cheaper than npm test, fails fast with a clear reason.
3. runtime/stuck-incident.renderActionTemplate (Voyager action_template.txt)
- All proposal bodies now follow the same fixed-section layout: Task /
Last result / Execution error / State / Metrics / Journal /
Scenarios / Critique / Fix / Edit scope / Forbidden.
4. runtime/skill-library.js (Mindcraft skill_library.getRelevantSkillDocs)
- Word-overlap ranking (Mindcraft's offline fallback) — zero deps,
deterministic. auto-patch.js injects top-3 similar skills into the
Pi prompt as "look at these patterns".
5. runtime/modes.js (Mindcraft modes.js)
- Declarative {name, interrupts, on, active, update(ctx)} chain that
runs BEFORE the curriculum each tick.
- Ships self_preservation (low HP → eat/flee), hunger (food<14 → eat),
night_shelter (night + bed in hand → sleep). Cleaner than ad-hoc
lastFleeAttempt cooldowns in reflex.js.
6. runtime/social/conversation.js + cmd:conv-say/conv-recent/conv-list
- File-JSONL topic channel so two bots from the same repo (different
usernames, different host dirs under state/) can append turns and
read peers. Skeleton — multi-agent collaboration on top later.
Differentiator preserved: every Pi-written skill still lands on main via
auto-patch.js (real git branch + smoke gate + cherry-pick). Voyager
keeps skills in a Chroma JSON, Mindcraft keeps them in RAM — pepa keeps
them as versioned source code reviewable in `git log`.
package.json: 0.0.1 → 0.1.0. 174/174 tests pass. README + AGENTS updated.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
mayatnikovandClaude Opus 4.7
|
d960db4819 |
feat(runtime): persistent memory — world-journal + scenario-memory
Closes a structural gap: the bot now actually REMEMBERS what it
discovered and what it tried. Two stores live under state/<host>/ and
are wired in automatically.
runtime/world-journal.js
- Append-only JSONL of discovered points (chopped, placed, base,
shelter, farm, dead_end). Indexed by 16-block spatial grid; O(neighbors)
nearest() lookups; 6 h age prune; 10k line ceiling with trim.
- leanestQuadrant({x,z}) reports the quadrant the bot has the FEWEST
markers in — used by explore.far to circle rather than retread.
- summary() exposed for the stuck-incident proposal body.
runtime/scenario-memory.js
- Sliding window of (skillId, situationHash, code, ok, detail) tuples.
- situationHash() is a coarse fingerprint (16x8x16 cell + day/night +
food/hp bucket + inv key set + closest hostile). So "same kind of
place + same kind of state" matches.
- shouldSkip({skillId, situation}) → true after ≥3 failures within 30
min UNLESS a more-recent success in the same situation un-locks it.
- recentTailFor() exposed for the stuck-incident body.
Wiring (runtime/bot.js):
- dispatchAction captures situationHash BEFORE the action runs and
records (skillId, situation, code, ok) after — failures are attributed
to the dispatch-time state, not the partial-effect state.
- worldDelta fields (choppedAt, minedAt, placedAt, baseAt, shelterAt,
plantedAt, harvestedAt, tilledAt) auto-flow into the journal.
- no_target + silent_dig_failure also write dead_end markers.
Scheduler / skills now consume memory:
- reflex.js curriculum reflex calls memory.shouldSkip — if the same
(skill, situation) failed 3+ times recently, auto-converts to a
wander hint so the bot leaves and tries elsewhere.
- explore.far calls journal.leanestQuadrant when multiple cardinal
directions are walkable and prefers the less-explored one.
- gather.logs walks to the nearest known "chopped" bucket within 96
blocks before falling through to findBlock — chunks with confirmed
trees are more likely to yield another.
stuck-incident body now includes journal byKind + last 12 scenario
entries so Pi can write a structural fix, not just a guard clause.
Architecturally: this is the foundation for "bot rewrites itself".
The proposals Pi now receives carry real signal about what was tried
and what's around, instead of a single snapshot in isolation.
10 new tests (world-journal × 5, scenario-memory × 5). npm test 134/134.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
mayatnikovandClaude Opus 4.7
|
19dc8e12c6 |
fix(runtime): unstick wander loop + chop radius + explore.far skill
Follow-up to the iteration-1 fixes. Live smoke on play.xmatic.team revealed the bot was spawning into a tree-less plain (no log within 32 blocks of spawn), looping wander→gather→no_target→wander forever inside a 16-block box. - runtime/actions.js: chopNearestTree search radius 32 → 64 (still no trees on this spawn, but a normal biome will be served well by it). wander now has a blind-walk fallback when pathfinder times out (look+forward+jump for 3 s) so the bot at least unsticks from leaves or pillars. Pathfinder timeout reduced 30 s → 15 s. - runtime/skills/explore-far.js: new explore.far skill — walks ~48 blocks in a quadrant (NE/SE/SW/NW, rotating per call) so successive hints actually circle the spawn instead of bouncing in place. Blind walk fallback included. - runtime/reflex.js: when the scheduler is told to wander twice in a row by gather.* recover hints, it now dispatches explore.far instead so the bot actually leaves the patch it's stuck in. Resets the consecutiveWanderHints counter on any success. - runtime/reflex.js (sleep): no longer dispatches when the bot has neither a bed in inventory NOR a known shelter/base location — saved one dispatch + 5-min cooldown per restart at night. - runtime/reflex.js (eat): inventory check + lastEatAt always updated fix the eat-spam loop observed live (every tick fired "eat" → "no food in inventory" → again). - runtime/skills/chop-logs.js: recognise "no log within ..." as no_target so the recover hint switches the bot to wander/explore. npm test 124/124. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
mayatnikovandClaude Opus 4.7
|
29542f0559 |
fix(runtime): unstick scheduler + chop + sleep + bed/shelter/farm skills
Recovers the bot from the live-server symptoms reported 2026-05-26: 1) constant supervisor reconnects, 2) chop "clicks once and stops", 3) sleep does nothing without a bed and so blocks night-skipping for other players, 4) curriculum reflex always fell through to wander. Supervisor (#38): - runtime/watch-filter.js: pure predicate excluding *.test.js + the supervisor itself; recursive:true so skills/ + social/ edits also restart. Burned a working main once when test files counted toward the rollback threshold. - runtime/supervisor.js: watch-triggered restarts no longer count toward the crash-loop rollback path. Watcher is now recursive. Chop / mine (#39): - runtime/actions.js + runtime/skills/gather-stone.js: replaced raw pathfinder.goto + bot.dig with mineflayer-collectblock's bot.collectBlock.collect — handles approach, repositioning, LoS, dig and pickup as one primitive. Old version "swung once" because GoalGetToBlock often parked the bot in leaves above the log. Sleep + bed (#40): - runtime/actions.js: sleepInBed now ALSO places a carried bed on solid ground next to the bot and sleeps on it. Critical so the bot stops blocking player night-skipping the moment it owns a bed. Bed pipeline (#41): - runtime/skills/gather-wool.js: gather.wool skill — mines wool block if any nearby, otherwise shears or attacks the nearest sheep. - runtime/skills/craft.js: craftBedSkill (any colour the bot has ≥3 wool of, plus 3 planks, plus a table). - runtime/curriculum.js: new milestone survive.bed sits between wood.tools and stone.32 so the bot gets a bed BEFORE everything else. Test fixture updated to include a red_bed in post-survive.bed stages. Village / shelter / wheat (#42, #43): - runtime/skills/build-shelter.js: village.build-shelter — real 3×3×3 resumable hut blueprint around the recorded base, places one block per loop, idempotent so an interrupted build resumes correctly, marks each placed block in the owned-blocks ledger. - runtime/skills/deposit-surplus.js: village.deposit-surplus opens the nearest chest and transfers surplus stacks while keeping a reserve of tools/food/bed. - runtime/skills/farm-wheat.js: farm.wheat does one step per call (till adjacent-to-water grass, plant seeds, or harvest ripe wheat). - runtime/curriculum.js: village.shelter milestone after base-site. Scheduler glitch (root of "always wander"): - runtime/bot.js: curriculum + locations are now computed BEFORE runTick. Previously they were stamped AFTER, so reflex.js saw snapshot.curriculum=undefined every tick and fell through to the wander fallback. Verified live: scheduler now dispatches gather.logs/gather.stone/craft.* by id via runSkill. Eat-spam: - runtime/reflex.js: eatReflex now checks inventory for actual food and updates lastEatAt on EVERY dispatch (not only successes), so a failed eat respects the 5 s cooldown instead of firing every tick. npm test 123/123. Validated live on play.xmatic.team (curriculum dispatched gather.logs via runSkill, recover hint switched to wander when no log in range). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
|
ea4f16a0da |
feat(runtime): scheduler-via-runSkill + Pi banter escalation + base-site (follow-ups) (#20)
Three closures of remaining PRD follow-ups, one merge:
1. Reflex scheduler now drives behaviour from the curriculum.
- reflex.js: replaced ad-hoc techTreeReflex + autonomousReflex with
curriculumReflex that dispatches the skill suggested by
snapshot.curriculum.plan via runSkill. Per-skill backoff for
missing_tool / missing_material / no_target / no_food_source /
unsupported_version. recover() hint with `{hint:"wander"}` swaps
the next tick to wander for 60 s.
- Chain is now: defend > eat > sleep > curriculum > idle.
- reflex.test.js: 11 new tests covering busy/disconnected,
defend/eat preemption, dispatch by id, unknown-skill fallback,
per-skill + wander-hint backoffs, onComplete updating backoff.
2. Pi escalation for ADDRESSED_BANTER with hard rate limit.
- bot.js: when generateReply returns {escalate:true}, spawn askPi
with bot state + last 5 lines from that speaker (redacted via
chatMemory). Reply capped at 200 chars, sent as one chat line.
- Rate cap: 6 calls/hour, 90 s min gap. Suppressed escalations
log once and silently drop.
3. Phase 4 substrate.
- runtime/locations.js: atomic JSON store
(state/<host>/locations.json) with setLocation / getLocation /
nearestLocation / removeLocation; 6 tests.
- runtime/base-site.js: scoreCurrentPosition(bot) + pure scoreSite
bundle (wood / stone / water / flatness / no-players /
no-foreign-builds, owned-blocks excluded from claim penalty);
6 tests.
- runtime/skills/choose-base.js: village.choose-base skill — scores
the current spot, writes locations.base if score ≥ 8, otherwise
returns code:"too_weak" with a wander recover hint.
- curriculum.js: new final milestone village.base-site fires
village.choose-base until a base location exists.
- bot.js: stamps snapshot.locations from listLocations() each tick
so the curriculum can read it without coupling to disk.
docs/runtime.md updated with three new sections.
npm test now 116/116.
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
|
3310cb320f |
feat(runtime): survival-bot pivot — MC chat is dialog-only (Phase 0) (#12)
Phase 0 of plans/autonomous-survival-bot-prd.md: change the product direction from operator-driven remote control to autonomous survival resident. MC chat is dialog-only for everyone, including OPERATOR_USERNAMES — commands like come/follow/build/pause/stop are recorded in the diary but not dispatched. TUI remains the only local control plane. Runtime changes: - Remove operatorGoalReflex from reflex.js (the come-here chat command). - Replace handleOperatorChat in bot.js with a dialog-only handleChat that answers greetings/status questions and records command-like verbs (en+ru) without dispatching them. - Default MC_VERSION to "auto" in runtime/config.js; mineflayer receives `false` to trigger version auto-detection. - Update auto-escalation prompt's reflex chain summary. Docs: - AGENTS.md: product pivot notice up top; chat-driven scope-trust is flagged as legacy/Pi-only. - README.md / docs/runtime.md: replace operator-chat command list with dialog-only description; update reflex chain summary. - docs/roadmap.md: Phase 2/3 marked superseded by the PRD where they assumed chat-driven control. Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
|
313ef081d1 |
feat(runtime): village progression — crafting, tech-tree reflex, LLM planner (#11)
Closes the loop "стой и кидай proposals" → "копит ресурсы, строит,
работает к глобальной цели". Three pieces:
1. Crafting primitives (runtime/actions.js).
craftPlanks (4 per log, any wood type), craftSticks (4 per 2 planks),
placeCraftingTable (crafts a table from planks if needed + places at
reference block + reuses an existing table within 4 m), craftWoodenAxe,
craftWoodenPickaxe, craftWoodenSword. Each uses bot.recipesFor()
+ bot.craft() with a 15s timeout. Returns the same {ok, detail}
contract as the other actions.
inv.{getItemCount, getAnyPlanksCount, getAnyLogCount} helpers
exported so the reflex layer can read inventory cheaply without
pulling mineflayer state through every reducer.
2. Tech-tree reflex (runtime/reflex.js).
New techTreeReflex between sleep and autonomous. Inventory-driven
progression: log+0 planks → planks; planks+0 sticks → sticks;
planks+sticks+no axe → wooden_axe; +no pickaxe → wooden_pickaxe;
+no sword → wooden_sword. 5 s cooldown so we don't fire on every
tick.
Pure script, no LLM. The progression is exactly what a player
does in the first 10 min on a new world; making it scripted means
the bot never burns tokens on it.
3. LLM planner (runtime/planner.js).
Background timer (every 15 min, with a 30 s warm-up after start).
Reads goal.md + plan.md + a slim snapshot, prompts Pi to output a
fresh plan.md to stdout. Stripped of code fences and written
verbatim to state/<host>/plan.md. Capped at 16 KB.
The plan is markdown the operator can read or edit by hand. Numbered
milestones, ✓ prefix for completed ones, kept short. The reflex
layer doesn't auto-execute LLM text — but the planner sets the
long-horizon shape that future reflexes (build house, plant farm)
can read.
5 min timeout on the pi subprocess. If it crashes or times out, the
next 15-min tick just retries — no propagation to the reflex loop.
The progression now looks like, roughly:
chop log (autonomous) →
craft planks → craft sticks → wooden_axe (tech-tree) →
chop faster (autonomous, has axe now) →
wooden_pickaxe + wooden_sword (tech-tree) →
mine stone … (next PR: stone tools, farm site selection,
house frame)
Smoke-tested: all three modules import cleanly, exports check out.
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
|
f5361ad9a4 |
feat(runtime): autonomous reflex — chop wood + wander + proactive movement (#8)
Closes the "bot stands on a tree doing nothing" problem reported live when the operator launched the TUI after PR #6 landed. The reactive chain (operator > defend > eat > sleep > idle) was passive by design: day-time, full HP and food, no hostile within 4 m ⇒ every reflex returned noop. The bot perched in dark-oak canopy and never moved. Changes runtime/actions.js: - chopNearestTree: find any *_log within 32 blocks, equip best axe (falls back to fists), path to the block, dig. Per-bot 5-min blacklist of unreachable log positions so we don't grind on the same impossible target. - wander: pick a random offset 6-16 blocks away and path there. - setMovementsForGather / setMovementsForTravel: every action that uses pathfinder now sets its own Movements profile (canDig=true) instead of inheriting whatever the previous caller left. The old behaviour caused chop to inherit flee's canDig=false and get stuck in the canopy. - fleeFrom now uses canDig=true too — the user observed the bot permanently stuck on a leaf block because escape required digging. runtime/reflex.js: - new autonomousReflex between sleep and idle. Cooldown 10s. Picks chop when log count < 16, else wander. When chop reports "no reachable log within 32 blocks" we switch to wander for 60s so we don't re-fire chop against the same impossible position. - defendReflex tightened: only flee when closest is ≤8m (or ≤12m on low HP). Avoids the "82 distant hostiles ⇒ constant flee loop" pathology observed at this spawn. - flee cooldown: same mob name within 60s ⇒ noop, so we yield to other reflexes if flee keeps timing out. - sleepReflex retry cooldown raised 30s → 5min. Sleeping fails permanently if no bed is around; the short retry blocked autonomous behaviour every tick. - ctx.lastReflex now records {name, label, ts} after each dispatched/completed reflex so the TUI can show what the bot just decided. runtime/bot.js: - per-tick snapshot adds lastReflex and busy fields for the TUI. - maybeReplyToPlayer: light canned greetings (yo/hey/hi/привет) to non-operators when they address the bot. 30s cooldown so we don't spam. tui/tui.tsx: - status bar shows either "▸ busy: <label>" while an action is in flight, or "last reflex: <name> (<label>) Ns ago" when idle. Gives an at-a-glance answer to "what is the bot doing right now?" Smoke-tested live (play.xmatic.team, 2026-05-25T13:30-13:39): - bot did dispatch chop tree (oak_log at 617,82,95) - real bug surfaced and 3-fail rule filed a proposal automatically - after the runtime fix, chop returns "no reachable log" gracefully - bot switched to wander on the next tick - position moved from (623.31, 85, 95.12) to (623.41, 86.02, 96.7) — the first observable movement in this dark-oak spawn Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
|
ee2b4c26eb |
feat(runtime): real reflex bodies + auto-escalation + operator chat (#5)
runtime/actions.js — Mineflayer wrappers with hard timeouts and structured
{ok, detail} returns:
- attackNearest: equip best melee, lookAt, single swing per call
- fleeFrom: lazy-load pathfinder, walk N blocks away (canDig=false to
avoid burrowing through walls under panic)
- eatBestFood: scan inventory by FOOD_PRIORITY, equip + consume
- sleepInBed: find nearest placed bed within 16 blocks, path to it, sleep
- goTo: pathfinder.goto for operator come/follow
runtime/reflex.js — bodies now dispatch real actions via ctx.dispatch:
- operator-goal (highest): satisfy come/follow command
- defend: ≤4m attack, ≤12m + low HP/many hostiles flee
- eat: food < 16 + 5s cooldown
- sleep: night + safe + 30s retry cooldown
- idle: heartbeat every 20th tick
Reflex returns "skipped" when ctx.busy so we don't count busy ticks as
either productive or noop in the escalation counter.
runtime/bot.js:
- ctx.dispatch fire-and-forget wrapper with busy gate, onComplete hook
- consecutiveNoops counter; after ESCALATE_AFTER_NOOPS (=20, ~1 min at
tick=3s), askPi with the current snapshot. 10-min cooldown.
- operator chat handler: parses `<botname> <verb>` messages from
OPERATOR_USERNAMES. Verbs: status, pause, resume, stop, come.
- Death drops any pending operator goal.
Smoke-tested live against play.xmatic.team:25565: bot connected, logged
in via AuthMe, reflex chain dispatched flee/sleep, hard timeout fired
when pathfinder couldn't reach the flee target (expected — no usable
ground path in dark_forest at this spawn).
Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
|
1e3b36a9a1 |
feat(runtime): hybrid script reflex + Ink TUI + Pi-on-demand escalation (#4)
* fix(mindcraft-skills): hard timeout on every skill call mc_avoid_enemies (and 7 other tools) wrapped only in safeCall without a withTimeout. When mindcraft's underlying pathfinder/pvp goal couldn't be satisfied, the call never resolved — the Pi tick loop blocked forever. Observed live: mc_avoid_enemies pending >10 minutes after one mc_observe. safeCall now takes timeoutMs (default 30s) and wraps withTimeout itself, so every tool gets a hard ceiling. Per-tool overrides: - goToPosition / goToNearestBlock: 120s / 90s (unchanged from before) - defendSelf / avoidEnemies: 45s - stay: secs*1000 + 10s - craft / consume / pickup / place: 30s - equip: 15s collectBlock still uses its bespoke per-iter 75s loop. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(runtime): script-driven reflex daemon + Ink TUI dashboard Pure-Pi runtime had three failure modes in practice: - slow: 20-60s per decision because LLM was in the hot path - expensive: every tick (defend, eat, idle) paid for a reasoning pass - invisible: required tmux capture-pane to know what the bot was doing New runtime/ layer is a long-running Node daemon that owns the MC connection, ticks a priority-ordered reflex chain (defend > eat > sleep > idle) with NO LLM in the hot path, and exposes status + commands over a Unix-socket IPC. tui/ is an Ink dashboard that attaches over IPC and can detach freely — multiple TUI clients can connect at once. Pi/Codex are still available, but as on-demand escalation: TUI hotkey 'a' spawns `pi -p "<prompt>"` as a subprocess and streams stdout into the dashboard. The self-improvement loop (proposals → operator approval → Pi-driven patch → hot reload) is documented in docs/runtime.md but not yet wired. Reflex bodies are stubs today — they log decisions but don't drive Mineflayer actions yet. The priority chain, IPC contract, and TUI are fully working; subsequent commits will fill in defend/eat/sleep bodies and wire automatic escalation. Run with `npm run bot` + `npm run tui`. Pi-only fallback stays at `npm run agent`. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Yuriy Mayatnikov <mayatnikov@me.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |