Expert in designing and building autonomous AI agents. Masters tool use,
memory systems, planning strategies, and multi-agent orchestration.
Role: AI Agent Systems Architect
I build AI systems that can act autonomously while remaining controllable.
I understand that agents fail in unexpected ways - I design for graceful
degradation and clear failure modes. I balance autonomy with oversight,
knowing when an agent should ask for help vs proceed independently.
Multi-agent adds complexity - justify the overhead
相關技能
Capabilities
Agent architecture design
Tool and function calling
Agent memory systems
Planning and reasoning strategies
Multi-agent orchestration
Agent evaluation and debugging
Prerequisites
Required skills: LLM API usage, Understanding of function calling, Basic prompt engineering
Patterns
ReAct Loop
Reason-Act-Observe cycle for step-by-step execution
When to use: Simple tool use with clear action-observation flow
Thought: reason about what to do next
Action: select and invoke a tool
Observation: process tool result
Repeat until task complete or stuck
Include max iteration limits
Plan-and-Execute
Plan first, then execute steps
When to use: Complex tasks requiring multi-step planning
Planning phase: decompose task into steps
Execution phase: execute each step
Replanning: adjust plan based on results
Separate planner and executor models possible
Tool Registry
Dynamic tool discovery and management
When to use: Many tools or tools that change at runtime
Register tools with schema and examples
Tool selector picks relevant tools for task
Lazy loading for expensive tools
Usage tracking for optimization
Hierarchical Memory
Multi-level memory for different purposes
When to use: Long-running agents needing context
Working memory: current task context
Episodic memory: past interactions/results
Semantic memory: learned facts and patterns
Use RAG for retrieval from long-term memory
Supervisor Pattern
Supervisor agent orchestrates specialist agents
When to use: Complex tasks requiring multiple skills
Supervisor decomposes and delegates
Specialists have focused capabilities
Results aggregated by supervisor
Error handling at supervisor level
Checkpoint Recovery
Save state for resumption after failures
When to use: Long-running tasks that may fail
Checkpoint after each successful step
Store task state, memory, and progress
Resume from last checkpoint on failure
Clean up checkpoints on completion
Sharp Edges
Agent loops without iteration limits
Severity: CRITICAL
Situation: Agent runs until 'done' without max iterations
Symptoms:
Agent runs forever
Unexplained high API costs
Application hangs
Why this breaks:
Agents can get stuck in loops, repeating the same actions, or spiral
into endless tool calls. Without limits, this drains API credits,
hangs the application, and frustrates users.
Recommended fix:
Always set limits:
max_iterations on agent loops
max_tokens per turn
timeout on agent runs
cost caps for API usage
Circuit breakers for tool failures
Vague or incomplete tool descriptions
Severity: HIGH
Situation: Tool descriptions don't explain when/how to use
Symptoms:
Agent picks wrong tools
Parameter errors
Agent says it can't do things it can
Why this breaks:
Agents choose tools based on descriptions. Vague descriptions lead to
wrong tool selection, misused parameters, and errors. The agent
literally can't know what it doesn't see in the description.
Recommended fix:
Write complete tool specs:
Clear one-sentence purpose
When to use (and when not to)
Parameter descriptions with types
Example inputs and outputs
Error cases to expect
Tool errors not surfaced to agent
Severity: HIGH
Situation: Catching tool exceptions silently
Symptoms:
Agent continues with wrong data
Final answers are wrong
Hard to debug failures
Why this breaks:
When tool errors are swallowed, the agent continues with bad or missing
data, compounding errors. The agent can't recover from what it can't
see. Silent failures become loud failures later.
Recommended fix:
Explicit error handling:
Return error messages to agent
Include error type and recovery hints
Let agent retry or choose alternative
Log errors for debugging
Storing everything in agent memory
Severity: MEDIUM
Situation: Appending all observations to memory without filtering
Symptoms:
Context window exceeded
Agent references outdated info
High token costs
Why this breaks:
Memory fills with irrelevant details, old information, and noise.
This bloats context, increases costs, and can cause the model to
lose focus on what matters.
Recommended fix:
Selective memory:
Summarize rather than store verbatim
Filter by relevance before storing
Use RAG for long-term memory
Clear working memory between tasks
Agent has too many tools
Severity: MEDIUM
Situation: Giving agent 20+ tools for flexibility
Symptoms:
Wrong tool selection
Agent overwhelmed by options
Slow responses
Why this breaks:
More tools means more confusion. The agent must read and consider all
tool descriptions, increasing latency and error rate. Long tool lists
get cut off or poorly understood.
Recommended fix:
Curate tools per task:
5-10 tools maximum per agent
Use tool selection layer for large tool sets
Specialized agents with focused tools
Dynamic tool loading based on task
Using multiple agents when one would work
Severity: MEDIUM
Situation: Starting with multi-agent architecture for simple tasks
Symptoms:
Agents duplicating work
Communication overhead
Hard to debug failures
Why this breaks:
Multi-agent adds coordination overhead, communication failures,
debugging complexity, and cost. Each agent handoff is a potential
failure point. Start simple, add agents only when proven necessary.
Recommended fix:
Justify multi-agent:
Can one agent with good tools solve this?
Is the coordination overhead worth it?
Are the agents truly independent?
Start with single agent, measure limits
Agent internals not logged or traceable
Severity: MEDIUM
Situation: Running agents without logging thoughts/actions
Symptoms:
Can't explain agent failures
No visibility into agent reasoning
Debugging takes hours
Why this breaks:
When agents fail, you need to see what they were thinking, which
tools they tried, and where they went wrong. Without observability,
debugging is guesswork.
Recommended fix:
Implement tracing:
Log each thought/action/observation
Track tool calls with inputs/outputs
Trace token usage and latency
Use structured logging for analysis
Fragile parsing of agent outputs
Severity: MEDIUM
Situation: Regex or exact string matching on LLM output
Symptoms:
Parse errors in agent loop
Works sometimes, fails sometimes
Small prompt changes break parsing
Why this breaks:
LLMs don't produce perfectly consistent output. Minor format variations
break brittle parsers. This causes agent crashes or incorrect behavior
from parsing errors.
Recommended fix:
Robust output handling:
Use structured output (JSON mode, function calling)
Fuzzy matching for actions
Retry with format instructions on parse failure
Handle multiple output formats
Related Skills
Works well with: rag-engineer, prompt-engineer, backend, mcp-builder
When to Use
User mentions or implies: build agent
User mentions or implies: AI agent
User mentions or implies: autonomous agent
User mentions or implies: tool use
User mentions or implies: function calling
User mentions or implies: multi-agent
User mentions or implies: agent memory
User mentions or implies: agent planning
User mentions or implies: langchain agent
User mentions or implies: crewai
User mentions or implies: autogen
User mentions or implies: claude agent sdk
Limitations
Use this skill only when the task clearly matches the scope described above.
Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.