Results for “stagnation-detection”

Trains, evaluates, exports, and runs inference for FastFoundationStereo (FFS) stereo depth estimation models, a distilled variant of FoundationStereo with lower latency.

Scan C# source files for hard-to-test static dependencies and produce a ranked report of static call sites by frequency.

Trains and fine-tunes vision models for object detection, image classification, and segmentation using Hugging Face Transformers on cloud GPUs, with automatic dataset validation and Hub persistence.

Diagnose and fix Vue 3 runtime errors, warnings, async failures, and SSR/hydration issues with targeted debugging guides.

Diagnoses and resolves slow Qdrant search performance issues including high latency, low throughput, and slow filtered searches.

Train, evaluate, distill, quantize, export, and run inference for RT-DETR object detection models using NVIDIA TAO.

Diagnoses Qdrant production issues using metrics and observability tools, covering optimizer problems, memory spikes, and slow queries.

Runs a fast quality gate that detects the project stack, performs static analysis, checks cross-layer consistency, and fixes issues between pipeline phases.

Detect and fix stale chunks (files that changed or were deleted since last indexing)

Runs a disciplined debugging loop for hard bugs and performance regressions, covering reproduction, hypothesis testing, instrumentation, fixing, and regression testing.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.

Detect data staging activity before exfiltration by monitoring for archive creation with 7-Zip/RAR, unusual temp folder access, large file consolidation, and staging directory patterns via EDR and process telemetry.

Investigates hard, flaky, or performance failures through a six-phase feedback-loop-first debugging process; use when quick targeted debugging is insufficient.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.

Performance regression detection using the browse daemon. Establishes baselines for page load times, Core Web Vitals, and resource sizes. Compares before/after on every PR. Tracks performance trends over time. Use when: "performance", "benchmark", "page speed", "lighthouse", "web vitals", "bundle size", "load time". (gstack) Voice triggers (speech-to-text aliases): "speed test", "check performance".

Detect and triage infrastructure drift by comparing declared Terraform state against live cloud resources using scheduled pipelines and audit logs.

Audit Archcore docs: dashboard (counts, status, relations, orphans), deep coverage audit, or drift detection (code/cascade/temporal staleness). Use for 'show status', 'documentation gaps', 'check if docs match code', or after a staleness warning. Not for creating docs.

Debugging constant Markov regime features in RL observations - when HMM probabilities show uniform values instead of dynamic regime estimates

The layers trade timeliness against definitiveness.

Design privacy-aware observability for AI agents using traces, spans, structured events, metrics, cost attribution, dashboards, alerts, and investigation workflows. Use when instrumenting an agent, debugging intermittent tool or model failures, defining service-level objectives, analyzing latency or spend, auditing agent decisions, or preparing production monitoring.

Detects DNS tunneling by computing Shannon entropy of DNS query names, analyzing query length distributions, inspecting TXT record payloads, and identifying high subdomain cardinality using scapy for packet capture analysis.

Pattern-Based Diagnosis

Diagnoses whether a user's current approach is trading short-term speed for long-term pain, and recommends slower methods that build compoundable assets through friction.

Manages automated dynamic/trailing stop losses (DSL only) for leveraged perpetual positions on Hyperliquid. Default mode: High Water (pct_of_high_water) — the trailing floor is a percentage of the peak ROE, recalculated every tick, no ceiling. Also supports fixed ROE tiers for legacy positions. Monitors price via cron, ratchets profit floors through configurable tiers, syncs the stop loss to Hyperliquid via edit_position, and auto-closes positions on breach via mcporter. Supports LONG and SHORT, strategy-scoped state isolation, and automatic cleanup.

Design loading, skeleton, and progressive content reveal patterns that maintain user confidence and perceived performance.

Runbook skill for failures: cron error, PA failure, cascade, gateway disconnect, semantic DB stale. Walks: detect → classify → diagnose → notify → log. Replaces ad-hoc failure handling. Triggers: "cron failed", "X is broken", "cascade", "incident", "gateway down", "PA failure".

HYDRA v2.0 — Squeeze Detector. Finds crowded trades about to unwind. Funding extreme + SM positioned against the crowd + price starting to move. Goes opposite to the funding crowd. Only liquid assets ($20M+ volume). DSL exit managed by plugin runtime via runtime.yaml.

Runs a disciplined diagnosis loop for hard bugs and performance regressions, from reproduction through hypothesis testing, instrumentation, fixing, and regression testing.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Diagnose failures from runtime evidence before editing code.

Use to measure whether a support bot, AI agent, or self-service channel actually reduces contact volume, and to audit a vendor's containment or deflection number. Trigger for "what's our real deflection rate", "is the bot working", "our containment rate is 70% but tickets haven't dropped", automation ROI, self-service savings, AI agent resolution rate, or designing a holdout test for a support bot.

Implements a complete observability stack with Prometheus metrics, Grafana dashboards, Jaeger distributed tracing, and structured logging, including heartbeat-based absence detection for scheduled jobs on Google Cloud.

Debugging is the scientific method under time pressure.

Resume an interrupted AgenTeam run with verify-first strategy.