Results for “tool-poisoning”

Apply least-privilege tool allowlisting, identity binding, and human-in-the-loop controls for agent tool calls.

Detect poisoned training data and backdoored models across the ML pipeline using statistical analysis, activation clustering, and spectral signatures.

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.

Conduct authorized security assessments of LLM applications and AI agents, covering prompt injection, tool abuse, RAG exposure, memory poisoning, and model supply-chain risks.

Provides a comprehensive penetration testing toolchain with 20+ security tools (Nmap, Nuclei, SQLMap, FFUF, Hashcat, etc.) exposed via MCP servers for authorized vulnerability scanning, exploitation, and reporting.

Probe RAG applications for prompt injection via poisoned retrieved context and embedding manipulation.

Hunt for threats in Windows Event Logs using Chainsaw, a fast Rust-based forensic tool that runs Sigma rules, keyword searches, and artifact analysis offline.

Design, run, and learn from safe chaos engineering experiments with structured plans, blast radius calculation, and postmortem generation.

Injects intentional failures to test system resilience using Litmus and Chaos Monkey, with a prerequisite to validate detection before running experiments.

Designs chaos experiments, creates failure injection frameworks, and facilitates game day exercises for distributed systems — producing runbooks, experiment manifests, rollback procedures, and post-mortem templates.

Investigate supply chain attack artifacts including trojanized software updates, compromised build pipelines, and sideloaded dependencies to identify intrusion vectors and scope of compromise.

Optional low-priority poller that inventories Worklink tool pins from <home>/worklink.yaml and files/reuses Chainlink bump issues when upstream versions drift. Opt-in: copy this directory into <home>/skills/worklink-tool-pins/ and configure tool_pins in worklink.yaml.

Detect and exploit JavaScript prototype pollution vulnerabilities on both client-side and server-side applications to achieve XSS, RCE, and authentication bypass through property injection.

A relentless interview to sharpen a plan or design.

Hunt for supply chain compromise indicators including trojanized software updates, compromised dependencies, unauthorized code modifications, and tampered build artifacts.

Design and validate model-facing tool definitions with clear names, action-oriented descriptions, bounded JSON Schema parameters, explicit side effects, safe defaults, idempotency, errors, and realistic tests. Use when creating function-calling tools, MCP tools, agent actions, structured tool inputs, or when a model selects the wrong tool, invents arguments, or causes unsafe side effects.

AI/ML security playbook. Use when assessing model supply chain attacks (pickle RCE, poisoned weights), adversarial examples, model poisoning, model stealing, data privacy attacks (membership inference, model inversion), and autonomous agent security risks.

Automates enrichment of raw indicators of compromise with multi-source threat intelligence context using SOAR platforms, Python pipelines, or TIP playbooks to reduce analyst triage time and standardize enrichment outputs.

Molecular machine learning toolkit. Property prediction (ADMET, toxicity), GNNs (GCN, MPNN), MoleculeNet benchmarks, pretrained models, featurization, for drug discovery ML.

Expert guidance for proactive threat hunting. Use this when the user asks to "hunt" for threads, IOCs, or specific TTPs.

Runs a relentless interview session to sharpen a plan or design.

Structured threat modeling skill using the PASTA framework (Process for Attack Simulation and Threat Analysis) combined with Adam Shostack's 4-question framework. Use this skill whenever the user asks to do threat modeling, security analysis, map the attack surface, identify threats, or review an application for security risks — even if they don't mention PASTA or a specific framework by name. Core activities: Component Mapping (architecture + data flows), Critical Assessment (business impact prioritization), and Logic Flaw Identification (attacker mindset on business logic). Produces: component map diagram (Mermaid), data flow diagram (Mermaid), attack tree (Mermaid), STRIDE threat table, prioritized risk register, and an actionable mitigation plan. Invoke proactively for any security review, architecture review, or "what could go wrong?" session.

Full-stack task execution pipeline — discovers tools via semantic search, routes to the right agent, then executes via GODMODE. Chains manus-tool-discovery → manus-orchestrator → godmode-system in sequence.

Detect process injection techniques (T1055) including CreateRemoteThread, process hollowing, and DLL injection via Sysmon Event IDs 8 and 10 and EDR process telemetry.

沈墨（悬疑剧虚构）认知与表达框架（压缩蒸馏）：创伤反杀叙事、时代灰雾、钢琴意象 触发：漫长的季节 等。虚构；禁止犯罪模仿

Plan, execute, document, and retest authorized security assessments of AI agents and multi-agent workflows using safe adversarial cases, synthetic identities, canaries, and evidence-based findings. Use when defining red-team rules of engagement, assessing prompt injection or excessive agency, testing tool and identity boundaries, evaluating memory or cross-agent attacks, scoring a campaign, or verifying remediation in an approved environment.

Use this skill for auth boundaries, input validation, secrets, RLS, injection risks, destructive-action review. Trigger when the task involves security work related to Security Hardening, production implementation, audits, debugging, strategy, or validation.

Read-only OWASP LLM Top 10 audit of app-facing AI features: prompt injection, data leak, supply chain, poisoning, unsafe output, excessive agency, system-prompt leak, RAG/embedding risks, misinformation, unbounded consumption. Use when "audit LLM security", "prompt injection", "jailbreak my chatbot", "is my AI safe".

Threat-model and harden AI agents, RAG systems, assistants, and tool-using workflows against direct, indirect, stored, cross-agent, and multimodal prompt injection. Use when reviewing an agent architecture, isolating untrusted content, constraining tools and egress, protecting secrets, adding injection-focused tests, investigating a suspected injection incident, or documenting residual prompt-injection risk.

Molecular machine learning toolkit. Property prediction (ADMET, toxicity), GNNs (GCN, MPNN), MoleculeNet benchmarks, pretrained models, featurization, for drug discovery ML.

Guides blue-team threat hunting and detection engineering with hypothesis-driven workflows, Sigma/YARA rule creation, SIEM query design, and validation using Atomic Red Team in authorized environments.

Use with analysis-agent, task-agent, or review-agent for task-local authentication lifecycle and recovery risk. Do not use without that decision or as task owner.

Review docs, tool output, skills, and memory candidates for prompt-injection risk.