Results for “model-backdoor”

Plan which model tier handles which work BEFORE execution begins — a high-cognition model deeply understands the problem, lays the foundations, then emits a modular plan assigning each module the cheapest tier that can safely execute it, with escalation tripwires and one-way-door protection. Advisory only: it announces "next module → tier X / model Y" at each boundary and the HUMAN switches models — harnesses like Cursor cannot switch mid-run. Load when the user asks which model to use, wants a model plan, model tiers, model-tier routing, assign models to tasks or modules, says "cheap model got stuck", "which model for this task", "cost-efficient model choice", or when implementation-plan / problem-to-plan need a model: tier column. NOT dynamic-routing (plan-path selection after failure) — this skill assigns cognition tiers to work.

Performs repository-grounded threat modeling by enumerating trust boundaries, assets, attacker capabilities, abuse paths, and mitigations, then writes a concise Markdown threat model.

Manage machine learning models in the Agent Platform Model Registry: list, describe, upload, update, and delete models and their versions.

Models threats for a service, feature, endpoint, integration, or architecture: assets, attackers, boundaries, flows, and abuse cases.

STRIDE Threat Model

Pixtral 12B: A Frontier Multimodal Model

Execute and analyze AI Data Poisoning attacks. By subtly injecting malicious or targeted misinformation into an LLM's training or fine-tuning dataset, an attacker can covertly manipulate the model's future outputs, implant backdoors, or enforce biases without altering the model architecture.

Packet-first BMAD/BMM front door for idea notes, product briefs, PRDs, architecture drafts, review feedback, existing repo state, and milestone pressure. Use when the user wants to know what BMAD phase or artifact comes next, or needs a portable BMAD entrypoint before routing review, execution slicing, runtime setup, or game-production work outward.

v5.6.0 joint multi-TF model: single model per symbol with broadcast 1Hour context replaces dual 15Min/1Hour models. Trigger: (1) replacing weighted-voting model aggregation, (2) adding broadcast features to vectorized env, (3) limited training data + worried about overfitting from doubling obs_dim, (4) backtest builder mismatch with newer feature counts.

Train machine learning models end-to-end, covering data loading, preprocessing, architecture selection, training loops, validation, and checkpointing. Use when the user requests model training or provides relevant inputs for this workflow.

Architecture specialist for model and training systems. Use when the user asks to talk to Chamber, requests the ML architect, or needs model architecture decisions.

Guides embedding model migration in Qdrant without downtime, covering alias swap, side-by-side, and hybrid search strategies.

Fine-tune open models or Gemini models using Agent Platform infrastructure, from environment setup through data preparation, job configuration, monitoring, and deployment.

Via modeling: pads, antipads, ground return vias, GRV placement, and signal integrity for high-speed layer transitions. TRIGGER: user asks to model a via, design a via transition, place ground return vias, analyze via performance, or check signal integrity through layer transitions. Invoke BEFORE writing code — only viaSingleEnded exists (no viaDifferential), and the location format is non-obvious. SKIP: general signal integrity without vias (use matlab-analyze-em), transmission line design (use matlab-design-pcb-transmission-line), PDN analysis (use matlab-analyze-pcb-pdn), material/stackup setup only (use matlab-manage-pcb-material).

Cross-model benchmark for gstack skills. (gstack)

Specialized agent for backend API development with REST/GraphQL endpoints, auth, database queries, and Controller-Service-Repository patterns