Results for “false-positive-tuning”

Use when tuning RL reward function weights, fixing reward hacking, or addressing P&L gradient issues

当在行动或追求目标过程中遇到挫折、痛苦或不满等负面体验时，调用此模型将负面体验转化为指引方向的积极信号

Triage mutation testing results by combining survived mutants, unnecessary test statements, and code graph analysis to identify false positives, missing test coverage, and fuzzing targets.

Hard Negative Mixing for Contrastive Learning

Detects Follow-Through Day (FTD) signals for market bottom confirmation using William O'Neil's methodology, with dual-index tracking and a state machine for rally attempt, FTD qualification, and post-FTD health monitoring.

Visual Instruction Tuning

当观察到或经历某种极端转化现象，或试图主动促成根本性观念转变时

Generates at most one token recommendation and one prediction-market pick per run, each with a falsifiable thesis, entry, sizing, and kill criterion. Returns NO_PICK when no candidate meets the bar.

Build clean async pipelines with fp-ts TaskEither, replacing nested try/catch with composable error handling, chaining, and parallel execution.

Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). (gstack)

Flip a problem, goal, or decision 180 degrees to find what forward thinking misses. Asks what would guarantee failure, then works backward to what must be avoided or changed. Load when the user says "invert this", "flip this problem", "what would guarantee failure", "think backward", "reverse engineer the goal", "think about it backwards", or when deep-thinking diagnoses an inversion frame. Also triggers on "what's the opposite of success here", "how would we sabotage this". Two methods: Failure Inversion and Opposite Goal. Max 2 clarifying questions before inverting. Always returns forward actions. For broader analysis, deep-thinking calls this.

Build clean async pipelines with fp-ts TaskEither, replacing nested try/catch with composable error handling and real API examples.

Performance-coaching mode for repeated failures, passive behavior, completion-quality issues, or explicit try-harder requests. Uses structured troubleshooting and evidence-first delivery habits.

Identifies common React useEffect anti-patterns and provides corrected implementations using derived state, event handlers, and cleanup functions.

Send a phone-push notification to the operator via ntfy.sh. Use ONLY for genuine algedonic alarms — events the operator needs to know about within minutes, not hours. The push lands on their phone; the cost of misuse is operator desensitization (they mute the topic and the channel goes dark). Do not use for routine surfacing, status updates, or things that can wait for the next chat turn.

Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). Review which AskUserQuestion prompts fire across gstack skills, set per-question preferences (never-ask / always-ask / ask-only-for-one-way), inspect the dual-track profile (what you declared vs what your behavior suggests), and enable/disable question tuning. Conversational interface — no CLI syntax required. Use when asked to "tune questions", "stop asking me that", "too many questions", "show my profile", "what questions have I been asked", "show my vibe", "developer profile", or "turn off question tuning". (gstack) Proactively suggest when the user says the same gstack question has come up before, or when they explicitly override a recommendation for the Nth time.

当遭遇挫折、失败或结果不如预期时，用于将负面经验转化为成长燃料

Deploy honeytokens and canarytokens as decoy artifacts to detect intrusions with near-zero false positives.

SSOT（Single Source of Truth）と実際のファイル/設定の整合性をチェックし、乖離があれば修正するスキル。 「SSOTチェックして」「SSOT整合性チェックして」「SSOT整理して」「SSOTのズレを直して」 「00_SYSTEM更新して」「乖離を修正して」と言った時にトリガーする。 /ssot-check でも呼び出せる。

Inverse ETF Proxy for Crypto Shorts

Meta-skill codifying the no-wakeup-polling pattern. Describes the ScheduleWakeup anti-pattern, the fix (--auto merge + detached watcher), and when to still use wakeups. Reduces handoff spam by 80%.

Performance-coaching mode for repeated failures, passive behavior, completion-quality issues, or explicit try-harder requests. Uses structured troubleshooting and evidence-first delivery habits.

Fix PPO training early-stop issues. Trigger when: (1) impossible drawdown values (>100%), (2) training stops too early, (3) need adaptive recovery instead of hard stop.

Build clean async pipelines with fp-ts TaskEither, replacing nested try/catch with composable error handling and parallel execution patterns.

Turn "is this account a real fit?" into a stacked verdict. A composite 0-100 fit score built from every signal you have, a breakdown by source, override flags for when a scoring tool is wrong, a channel classification, and a written rationale. Built for reps, marketers, and RevOps, customizable to your CRM and whatever scoring you use. Trigger on "is {account} a real fit?", "score this list against our profile", "find lookalikes to our best customers", "why is the score wrong on {account}?", "validate this prospect list", "what's our coverage in {segment}?", or any account or list qualification.

Fix 8-action space bug in backtest engines and rewrite backtest notebook for Colab v4.0.0

Reduces SOC alert fatigue by tuning detection rules, consolidating duplicate alerts, implementing risk-based alerting, and measuring alert quality metrics to maintain analyst effectiveness.