* 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>
246 lines
9.0 KiB
JavaScript
246 lines
9.0 KiB
JavaScript
// Auto-trigger policy for the fast tactical advisor.
|
|
//
|
|
// The advisor is too slow for synchronous use inside a reflex tick
|
|
// (5-15s via TimeWeb's hosted agent endpoint). The strategy here is
|
|
// asynchronous: when conditions warrant tactical advice, fire-and-forget
|
|
// an advise() call; when the result eventually arrives, cache it on
|
|
// ctx.advisorRecommendation. The next reflex tick reads that cache and
|
|
// can substitute the recommended skill before dispatching.
|
|
//
|
|
// Triggers (any one, AND-ed with the not-recently-asked cooldown):
|
|
//
|
|
// 1. Wedged > 60s — bot's position hasn't shifted ≥16 blocks in over
|
|
// a minute (already tracked by reflex.js as lastSignificantMoveAt)
|
|
// 2. Last 4+ dispatches were the same skill — clear loop signal
|
|
// 3. Last awareness preempt was very recent AND followed by same
|
|
// skill being dispatched again — env-shock-blind retry
|
|
//
|
|
// The recommendation has a TTL (60s). After that it's stale and the
|
|
// reflex falls back to manifesto / curriculum. This keeps the system
|
|
// reactive — advice ages out, fresh data drives fresh advice.
|
|
|
|
import { advise, isAvailable as advisorAvailable } from "./fast-advisor.js";
|
|
import { isRegistered } from "../skill-registry.js";
|
|
import { insertRecommendation } from "../knowledge/index.js";
|
|
import { info, warn } from "../log.js";
|
|
|
|
const TRIGGER_COOLDOWN_MS = 90_000;
|
|
const RECOMMENDATION_TTL_MS = 60_000;
|
|
const WEDGED_THRESHOLD_MS = 60_000;
|
|
const REPEAT_THRESHOLD = 4;
|
|
const PREEMPT_WINDOW_MS = 30_000;
|
|
// Emergency triggers — bypass cooldown because waiting another 90s
|
|
// when the bot is about to die is not useful.
|
|
const EMERGENCY_HP = 6;
|
|
const EMERGENCY_HOSTILE_DIST = 8;
|
|
const EMERGENCY_COOLDOWN_MS = 20_000;
|
|
// LLM outage backoff — if 3 consecutive advise() calls return
|
|
// http_400 / network_error, suppress further calls for 10 minutes.
|
|
const PROVIDER_OUTAGE_FAILS = 3;
|
|
const PROVIDER_OUTAGE_BACKOFF_MS = 10 * 60 * 1000;
|
|
|
|
let _lastTriggerAt = 0;
|
|
let _inFlight = false;
|
|
let _consecutiveFails = 0;
|
|
let _providerOutageUntil = 0;
|
|
|
|
export function _resetForTest() {
|
|
_lastTriggerAt = 0;
|
|
_inFlight = false;
|
|
_consecutiveFails = 0;
|
|
_providerOutageUntil = 0;
|
|
}
|
|
|
|
function isProviderError(code) {
|
|
return code === "network_error"
|
|
|| code === "timeout"
|
|
|| (typeof code === "string" && code.startsWith("http_"));
|
|
}
|
|
|
|
export function getTriggerState() {
|
|
return {
|
|
lastTriggerAt: _lastTriggerAt,
|
|
inFlight: _inFlight,
|
|
};
|
|
}
|
|
|
|
/**
|
|
* tickAdvisor(ctx) → maybe-fires advise() in background.
|
|
*
|
|
* Called from reflex AFTER it has chosen a plannedSkillId but BEFORE
|
|
* dispatching. Does NOT block — the in-flight call resolves later and
|
|
* writes ctx.advisorRecommendation. The caller decides whether to
|
|
* consume a fresh recommendation on this tick or wait for the next.
|
|
*/
|
|
export function tickAdvisor(ctx, { plannedSkillId } = {}) {
|
|
if (!advisorAvailable()) return { fired: false, reason: "disabled" };
|
|
if (_inFlight) return { fired: false, reason: "in_flight" };
|
|
|
|
const now = Date.now();
|
|
if (_providerOutageUntil > now) {
|
|
return { fired: false, reason: "provider_outage", retryAt: _providerOutageUntil };
|
|
}
|
|
|
|
// Drop a recommendation that's already aged out.
|
|
if (ctx.advisorRecommendation && now - ctx.advisorRecommendation.at > RECOMMENDATION_TTL_MS) {
|
|
ctx.advisorRecommendation = null;
|
|
}
|
|
|
|
const reason = detectTrigger(ctx, now, plannedSkillId);
|
|
if (!reason) return { fired: false, reason: "no_trigger" };
|
|
|
|
// Emergency triggers use a much shorter cooldown — waiting 90s with
|
|
// HP=4 and a creeper at 3 blocks is exactly when we MUST hit the LLM.
|
|
const isEmergency = reason.startsWith("emergency_");
|
|
const cooldownMs = isEmergency ? EMERGENCY_COOLDOWN_MS : TRIGGER_COOLDOWN_MS;
|
|
if (now - _lastTriggerAt < cooldownMs) {
|
|
return { fired: false, reason: "cooldown" };
|
|
}
|
|
|
|
_lastTriggerAt = now;
|
|
_inFlight = true;
|
|
const snapshot = ctx.snapshot ?? null;
|
|
const recentSkillIds = (ctx.recentSkillIds ?? []).slice(-8);
|
|
const activeNeed = ctx.activeNeed ?? null;
|
|
const storyStep = ctx.storyStep ?? null;
|
|
|
|
info("advisor-trigger", `firing because ${reason} (planned=${plannedSkillId ?? "?"}, need=${activeNeed?.need?.id ?? "?"}, step=${storyStep?.step?.id ?? "?"})`);
|
|
// Fire-and-forget. The promise's resolution writes ctx.advisorRecommendation.
|
|
advise({ snapshot, reason, recentSkillIds, lessonsTail: ctx.recentLessons ?? [], activeNeed, storyStep, force: true })
|
|
.then((result) => {
|
|
_inFlight = false;
|
|
const needLabel = activeNeed
|
|
? `L${activeNeed.need.level} ${activeNeed.need.id}`
|
|
: null;
|
|
if (result.ok && result.action === "switch_skill" && isRegistered(result.skillId)) {
|
|
const recId = insertRecommendation({
|
|
triggerReason: reason,
|
|
plannedSkill: plannedSkillId ?? null,
|
|
recommendedSkill: result.skillId,
|
|
action: "switch_skill",
|
|
rationale: result.rationale,
|
|
activeNeed: needLabel,
|
|
tokensIn: result.usage?.in,
|
|
tokensOut: result.usage?.out,
|
|
latencyMs: result.latencyMs,
|
|
});
|
|
ctx.advisorRecommendation = {
|
|
id: recId,
|
|
at: Date.now(),
|
|
skillId: result.skillId,
|
|
action: "switch_skill",
|
|
rationale: result.rationale,
|
|
triggerReason: reason,
|
|
latencyMs: result.latencyMs,
|
|
usage: result.usage ?? null,
|
|
};
|
|
info("advisor-trigger", `recommendation cached: ${result.skillId} (${result.latencyMs}ms, in=${result.usage?.in ?? "?"}t/out=${result.usage?.out ?? "?"}t, db=${recId ?? "-"})`);
|
|
} else if (result.ok && (result.action === "wait" || result.action === "continue")) {
|
|
const recId = insertRecommendation({
|
|
triggerReason: reason,
|
|
plannedSkill: plannedSkillId ?? null,
|
|
recommendedSkill: null,
|
|
action: result.action,
|
|
rationale: result.rationale,
|
|
activeNeed: needLabel,
|
|
tokensIn: result.usage?.in,
|
|
tokensOut: result.usage?.out,
|
|
latencyMs: result.latencyMs,
|
|
});
|
|
ctx.advisorRecommendation = {
|
|
id: recId,
|
|
at: Date.now(),
|
|
action: result.action,
|
|
rationale: result.rationale,
|
|
triggerReason: reason,
|
|
latencyMs: result.latencyMs,
|
|
usage: result.usage ?? null,
|
|
};
|
|
info("advisor-trigger", `recommendation: ${result.action} (${result.latencyMs}ms)`);
|
|
} else if (!result.ok) {
|
|
warn("advisor-trigger", `advise failed: ${result.code} (${result.detail})`);
|
|
if (isProviderError(result.code)) {
|
|
_consecutiveFails += 1;
|
|
if (_consecutiveFails >= PROVIDER_OUTAGE_FAILS) {
|
|
_providerOutageUntil = Date.now() + PROVIDER_OUTAGE_BACKOFF_MS;
|
|
warn("advisor-trigger", `LLM provider outage (${_consecutiveFails} fails in a row); backing off ${Math.round(PROVIDER_OUTAGE_BACKOFF_MS / 60000)}min`);
|
|
}
|
|
}
|
|
} else {
|
|
_consecutiveFails = 0;
|
|
}
|
|
})
|
|
.catch((e) => {
|
|
_inFlight = false;
|
|
warn("advisor-trigger", `advise threw: ${e?.message ?? e}`);
|
|
});
|
|
|
|
return { fired: true, reason };
|
|
}
|
|
|
|
function detectTrigger(ctx, now, plannedSkillId) {
|
|
const snap = ctx.snapshot ?? {};
|
|
|
|
// 0. EMERGENCY: low HP + hostile near — call BEFORE the bot dies.
|
|
// Checked first so reason string starts with "emergency_" → bypasses
|
|
// the long trigger cooldown via the caller's isEmergency check.
|
|
const hp = snap.health ?? 20;
|
|
const hostile = snap.closestHostile;
|
|
if (hp <= EMERGENCY_HP && hostile && (hostile.distance ?? Infinity) <= EMERGENCY_HOSTILE_DIST) {
|
|
return `emergency_hp${Math.round(hp)}_${hostile.name ?? "hostile"}@${Math.round(hostile.distance)}`;
|
|
}
|
|
// 0b. EMERGENCY: drowning / lava / lethal fluid
|
|
if (snap.hazards?.footBlock === "lava") {
|
|
return "emergency_lava";
|
|
}
|
|
|
|
// 1. Wedged > threshold
|
|
if (ctx.lastSignificantMoveAt && (now - ctx.lastSignificantMoveAt) > WEDGED_THRESHOLD_MS) {
|
|
return `wedged_${Math.round((now - ctx.lastSignificantMoveAt) / 1000)}s`;
|
|
}
|
|
// 2. Repeat-skill loop
|
|
const recent = ctx.recentSkillIds ?? [];
|
|
if (recent.length >= REPEAT_THRESHOLD) {
|
|
const tail = recent.slice(-REPEAT_THRESHOLD);
|
|
const allSame = tail.every((id) => id === tail[0]);
|
|
if (allSame && plannedSkillId === tail[0]) {
|
|
return `repeat_${REPEAT_THRESHOLD}_${tail[0]}`;
|
|
}
|
|
}
|
|
// 3. Recent preempt followed by same skill again
|
|
if (ctx.lastPreempt && now - ctx.lastPreempt.at < PREEMPT_WINDOW_MS) {
|
|
const lastDispatched = recent[recent.length - 1];
|
|
if (lastDispatched && lastDispatched === plannedSkillId) {
|
|
return `preempt_retry_${ctx.lastPreempt.reason}`;
|
|
}
|
|
}
|
|
return null;
|
|
}
|
|
|
|
/**
|
|
* consumeFreshRecommendation(ctx) → { skillId, action, rationale } | null
|
|
*
|
|
* Returns a recommendation if one is currently cached and fresh, and
|
|
* clears it so subsequent ticks don't re-apply the same advice.
|
|
*/
|
|
export function consumeFreshRecommendation(ctx) {
|
|
const rec = ctx.advisorRecommendation;
|
|
if (!rec) return null;
|
|
if (Date.now() - rec.at > RECOMMENDATION_TTL_MS) {
|
|
ctx.advisorRecommendation = null;
|
|
return null;
|
|
}
|
|
if (rec.action !== "switch_skill" || !rec.skillId) {
|
|
// 'continue' / 'wait' don't replace dispatch; surface for telemetry only
|
|
return null;
|
|
}
|
|
ctx.advisorRecommendation = null;
|
|
return rec;
|
|
}
|
|
|
|
// Test exports
|
|
export const __testing = {
|
|
TRIGGER_COOLDOWN_MS, RECOMMENDATION_TTL_MS, WEDGED_THRESHOLD_MS,
|
|
REPEAT_THRESHOLD, PREEMPT_WINDOW_MS, detectTrigger,
|
|
};
|