Tencent
Official@tencent · Shenzhen, China
Offers specialized security auditing, vulnerability detection, and data processing capabilities for enterprise environments and large-scale digital infrastructure.
Agent Skills by Tencent
Showing 33 vetted skills indexed across 7 GitHub repositories.
Tencentvery_simple_fela
Generates and evaluates feature engineering code for AUC maximization on tabular benchmarks.
Tencentmas-compiler
Compiles multi-agent systems into single-agent skills via three-phase pipeline compression.
Tencentvery_simple_apex
Generates SQLite queries from natural language questions using schema exploration, curated SQL tips, and reward-model ranking.
Tencentadaptive_apex_spider
Generates and validates Snowflake SQL from natural-language questions on Spider 2.0-Snow databases.
Tencentadaptive_mas_converter
Convert multi-agent systems into single-agent skills based on evaluation metric freedom.
Tencentbrowser-skill
Automate a logged-in Chromium browser through the bsk CLI for navigation, form filling, and data extraction.
Tencentmemory-poisoning-detection
Detect persistent instruction injection in agent long-term memory through dialogue probes.
Tencentweb-exfiltration-detection
Detect data exfiltration via URL path encoding and chained web navigation in AI agents.
Tencentcascading-failure-detection
Detect cascading failures and error propagation in multi-step agent workflows.
Tencentfile-path-traversal-detection
Detects path traversal vulnerabilities in agent file upload, save, and extract flows.
Tencentagentic-supply-chain-detection
Detect supply-chain compromise risks in AI agents via dialogue-based probing.
Tencenthuman-agent-trust-exploit-detection
Detect social engineering and deceptive guidance in agent dialogue responses.
Tencentunexpected-code-execution-detection
Detect command injection, eval/exec abuse, and remote code loading in AI agents.
Tencentdirect-injection-detection
Detects direct prompt injection and role override attempts in agent dialogue responses.
Tencentinter-agent-comm-security-detection
Detect data leakage, boundary bypass, and privilege mismatch in inter-agent communication.
Tencenthardcoded-secret-detection
Detect hardcoded secrets in agent-accessible source code and configuration files.
Tencentaig-agent-redteam
Conduct authorized red-team security assessments of AI agents, MCP servers, and AI infrastructure.
Tencentedgeone skill scanner
Audit agent skills for security risks using local static analysis.
Tencentweknora-shared
Operate a WeKnora RAG server through the weknora CLI with JSON envelopes and typed exit codes.
Tencentweknora-rag-search
Select and run the correct WeKnora CLI command for knowledge base retrieval and RAG queries.
Tencent文档协作
Guides collaborative document creation through context gathering, iterative drafting, and reader testing.
Tencentworkrally
Manage AIGC video production workflows via a CLI toolset.
Tencentbrowser-skill
Automate login, navigation, form filling, and data scraping in Chromium.
Tencentyolo-master-agent
Orchestrate YOLO-Master training, validation, and prediction workflows with Ultralytics.
Frequently Asked Questions About Tencent
FAQPage SchemaWhat security tasks can be performed using these capabilities?▼
These capabilities enable the classification of security findings using the OWASP ASI 2026 framework, detection of command injection, identification of sensitive information disclosure, and assessment of authorization bypass vulnerabilities through simulated privilege escalation and cross-user access testing.
Which technical personas benefit from these security modules?▼
Security engineers, penetration testers, and compliance officers benefit from these modules. They are designed for professionals tasked with hardening AI-integrated systems, auditing supply chain configurations, and ensuring data privacy across complex enterprise digital environments.
What are the primary prerequisites for deploying these security checks?▼
Deployment requires an environment configured for dialogue-based interaction testing and access to the target systems being audited. Users must ensure their infrastructure supports the specific detection modules, such as OpenClaw for configuration audits or Chromium-based environments for browser-related data scraping tasks.