Browse
Search and filter by type across the catalog
1,517 results in Identity, Sub-Agents, Memory, Workflows · page 34 of 64
legal-advisor-2
Use this agent when you need to draft contracts, review compliance requirements, develop IP protection strategies, or assess legal risks for technology businesses. Specifically:\\n\\n<example>\\nContext: Startup is launching a SaaS product and needs to establish foundational legal protection before customer agreements.\\nuser: \"We need a Terms of Service, Privacy Policy, and GDPR compliance roadmap for our new SaaS product launching in 3 months. We're handling user data and need to ensure we're legally protected.\"\\nassistant: \"I'll develop a comprehensive legal framework: draft compliant ToS with liability limitations, create a Privacy Policy addressing GDPR and CCPA requirements, establish data processing procedures, design consent flows, and provide a compliance checklist with implementation timeline. I'll also identify key jurisdictions to address and potential gaps in your current data handling.\"\\n<commentary>\\nUse legal-advisor when launching products or services that require legal infrastructure like ToS, privacy policies, or data handling compliance. This covers multi-jurisdictional requirements and proactive legal framework setup.\\n</commentary>\\n</example>\\n\\n<example>\\nContext: Company is signing significant vendor contracts and needs risk assessment before commitment.\\nuser: \"We're evaluating a major cloud infrastructure contract with AWS. Can you review this agreement and identify risky clauses, liability exposures, and negotiation points? We want to understand what we're signing up for.\"\\nassistant: \"I'll conduct a detailed contract analysis: identify liability caps and indemnification issues, flag unclear SLA terms, assess penalty clauses, review data ownership and security requirements, highlight auto-renewal and termination provisions, and prioritize negotiation points by risk level. I'll provide specific recommended language changes and fallback positions.\"\\n<commentary>\\nInvoke legal-advisor when reviewing or negotiating vendor contracts, partnership agreements, or other binding commitments. This focuses on protecting business interests while identifying negotiable terms.\\n</commentary>\\n</example>\\n\\n<example>\\nContext: Tech company wants to strengthen IP protection and avoid infringement risks.\\nuser: \"We need to audit our intellectual property strategy. We've built proprietary algorithms and tools, and we want to understand: should we patent, what trade secrets need protecting, do we need trademark registration? Also checking if we're infringing anything.\"\\nassistant: \"I'll develop a comprehensive IP strategy: assess patentability of your algorithms, recommend trademark registration approach for your brand and tools, establish trade secret protection procedures, create employee IP assignment policies, conduct competitive analysis to identify infringement risks, and propose licensing agreements for any third-party dependencies.\"\\n<commentary>\\nUse legal-advisor for intellectual property strategy when you need to protect proprietary technology, establish trademark/patent strategy, or assess infringement risks. This is critical before product launch or significant funding rounds.\\n</commentary>\\n</example>
llm-redteam-specialist
Use this agent when you need to red-team a Large Language Model deployment — jailbreak probes, prompt injection harness design, output-safety evaluation, and robustness evidence for EU AI Act Article 15 or NIST AI RMF MEASURE-2.7. Covers cloud-hosted models and on-prem / air-gapped local models (Ollama, vLLM, llama.cpp). Specifically:\\n\\n<example>\\nContext: A healthcare vendor embeds an LLM in a clinical triage tool and the compliance team wants a red-team report before rollout.\\nuser: \"We're deploying a Llama-3 70B behind a clinical assistant. Legal wants evidence it won't hand out harmful medical advice or leak PHI from retrieval context. How do we test it and document it?\"\\nassistant: \"I'll design an air-gapped red-team harness: a probe suite covering jailbreak families (DAN, role-play escalation, encoding attacks, prompt-leaking, indirect injection via retrieved docs), a scoring rubric aligned to the deployment's harm taxonomy, and a repeatable runner targeting your Ollama endpoint. Output is a robustness evidence pack: pass/fail table, example transcripts, coverage metric, and a control narrative mapped to NIST AI RMF MEASURE-2.7 and EU AI Act Article 15.\"\\n<commentary>\\nInvoke llm-redteam-specialist when the question is about evaluating an LLM's resistance to adversarial input — not generic web pentesting. This agent understands jailbreak taxonomies and the difference between a model-level test and a system-level test.\\n</commentary>\\n</example>\\n\\n<example>\\nContext: A defense contractor is running only local models (no external API calls allowed) and needs offline evaluation tooling.\\nuser: \"Air-gapped network. No HuggingFace, no OpenAI API. We still need quarterly robustness evidence for Llama-3 and Mistral instances. Build the test rig.\"\\nassistant: \"I'll spec an offline harness: probe corpus committed to the local repo, runner that targets localhost Ollama / vLLM endpoints, deterministic scoring (no model-as-judge calls outside the enclave), and a signed evidence bundle per run. Retention and signing align to the site's audit requirements. I'll pair this with a schedule for re-runs after every model or system-prompt change.\"\\n<commentary>\\nUse when the environment forbids cloud-hosted grader models and probe corpora must be self-contained.\\n</commentary>\\n</example>\\n\\n<example>\\nContext: A SaaS company received a prospect security questionnaire asking for evidence of prompt-injection testing.\\nuser: \"Enterprise prospect wants evidence we test for prompt injection. What do I send them?\"\\nassistant: \"I'll produce a prompt-injection test report: scope (which endpoints and retrieval paths were tested), probe inventory with OWASP LLM Top 10 references, results table, severity rubric, and remediation status per finding. I'll also flag the gap between direct-injection and indirect-injection coverage so the evidence is honest.\"\\n<commentary>\\nInvoke for LLM-specific adversarial evidence — distinct from penetration-tester which covers web/network.\\n</commentary>\\n</example>
llms-maintainer
LLMs.txt roadmap file generator and maintainer for AI Engine Optimization (AEO). Use after build completion, content changes, or when setting up AI crawler navigation for a site. Detects framework, scans site structure, and writes a spec-compliant llms.txt file.
load-testing-specialist
Load testing and stress testing specialist. Use PROACTIVELY for creating comprehensive load test scenarios, analyzing performance under stress, and identifying system bottlenecks and capacity limits.
m365-admin
Use when automating Microsoft 365 administrative tasks including Exchange Online mailbox provisioning, Teams collaboration management, SharePoint site configuration, license lifecycle management, and Graph API-driven identity automation.
malware-analyst
Expert malware analyst specializing in defensive malware research, threat intelligence, and incident response. Masters sandbox analysis, behavioral analysis, and malware family identification. Handles static/dynamic analysis, unpacking, and IOC extraction. Use PROACTIVELY for malware triage, threat hunting, incident response, or security research.
managed-agents-cma-coordinate-specialist-team
Heterogeneous team via the multiagent coordinator config — a coordinator runs three specialists (web-search researcher, file-reading librarian, rules-based pricer) with scoped toolsets to assemble a sales proposal. Covers the multiagent field, the thread_created / thread_message_received event types, and per-role to…
managed-agents-cma-iterate-fix-failing-tests
Entry-point tutorial for the Claude Managed Agents API. Walks through agent / environment / session creation, file mounts, and the streaming event loop by getting an agent to fix three planted bugs in a calc.py package.
managed-agents-cma-operate-in-production
End-to-end production story for Managed Agents — vault-backed MCP credentials, the session.status_idled webhook pattern for human-in-the-loop without long-lived connections, and the resource lifecycle CRUD verbs.