Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

AI Agent Development Workflow · sickn33 | DecimalAI

AI Agent Development

SkillScore77Maintained

ai-agent-development

AI agent development workflow for building autonomous agents, multi-agent systems, and agent orchestration with CrewAI, LangGraph, and custom agents.

agent-toolingby sickn33

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

AI Engineer pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/ai-agent-development/SKILL.md

decimalai skills pull ai-agent-development --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/ai-agent-development/SKILL.md https://app.decimal.ai/s/ai-agent-development/SKILL.md

Raw file: /s/ai-agent-development/SKILL.md ↗ · machine-readable summary at /s/ai-agent-development.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 60d ago

—-16% tokens0% turns77% of whole cases pass with skill (17/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-22	✗→✓	▲ Improved	—	—
case-03	✗→✓	▲ Improved	—	—
case-04	✗→✓	▲ Improved	—	—
case-08	✗→✓	▲ Improved	—	—
case-05	✗→✓	▲ Improved	—	—

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/30/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

sickn33 / antigravity-awesome-skills

skills/ai-agent-development/SKILL.md

View on GitHub ↗

License MITImported 8/2/2026Version v1

Skill contents

SKILL.md

AI Agent Development Workflow

Overview

Specialized workflow for building AI agents including single autonomous agents, multi-agent systems, agent orchestration, tool integration, and human-in-the-loop patterns.

When to Use This Workflow

Use this workflow when:

Building autonomous AI agents
Creating multi-agent systems
Implementing agent orchestration
Adding tool integration to agents
Setting up agent memory

Workflow Phases

Phase 1: Agent Design

Skills to Invoke

ai-agents-architect - Agent architecture
autonomous-agents - Autonomous patterns

Actions

Define agent purpose
Design agent capabilities
Plan tool integration
Design memory system
Define success metrics

Copy-Paste Prompts

Use @ai-agents-architect to design AI agent architecture

Phase 2: Single Agent Implementation

Skills to Invoke

autonomous-agent-patterns - Agent patterns
autonomous-agents - Autonomous agents

Actions

Choose agent framework
Implement agent logic
Add tool integration
Configure memory
Test agent behavior

Copy-Paste Prompts

Use @autonomous-agent-patterns to implement single agent

Phase 3: Multi-Agent System

Skills to Invoke

crewai - CrewAI framework
multi-agent-patterns - Multi-agent patterns

Actions

Define agent roles
Set up agent communication
Configure orchestration
Implement task delegation
Test coordination

Copy-Paste Prompts

Use @crewai to build multi-agent system with roles

Phase 4: Agent Orchestration

Skills to Invoke

langgraph - LangGraph orchestration
workflow-orchestration-patterns - Orchestration

Actions

Design workflow graph
Implement state management
Add conditional branches
Configure persistence
Test workflows

Copy-Paste Prompts

Use @langgraph to create stateful agent workflows

Phase 5: Tool Integration

Skills to Invoke

agent-tool-builder - Tool building
tool-design - Tool design

Actions

Identify tool needs
Design tool interfaces
Implement tools
Add error handling
Test tool usage

Copy-Paste Prompts

Use @agent-tool-builder to create agent tools

Phase 6: Memory Systems

Skills to Invoke

agent-memory-systems - Memory architecture
conversation-memory - Conversation memory

Actions

Design memory structure
Implement short-term memory
Set up long-term memory
Add entity memory
Test memory retrieval

Copy-Paste Prompts

Use @agent-memory-systems to implement agent memory

Phase 7: Evaluation

Skills to Invoke

agent-evaluation - Agent evaluation
evaluation - AI evaluation

Actions

Define evaluation criteria
Create test scenarios
Measure agent performance
Test edge cases
Iterate improvements

Copy-Paste Prompts

Use @agent-evaluation to evaluate agent performance

Agent Architecture

User Input -> Planner -> Agent -> Tools -> Memory -> Response
              |          |        |        |
         Decompose   LLM Core  Actions  Short/Long-term

Quality Gates

] Agent logic working
] Tools integrated
] Memory functional
] Orchestration tested
] Evaluation passing

Related Workflow Bundles

ai-ml - AI/ML development
rag-implementation - RAG systems
workflow-automation - Workflow patterns

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.

Illustrative Want to see what ai-agent-development would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

7/30/2026, 08:16 UTC · 22 cases

Measured on 21 comparable cases — below the 22 needed to publish a lift, so this result is shown here but not on the registry card or leaderboard.

Lift

+77%

77% vs 0%

Agent turns

0%

1.0 vs 1.0

Tokens

-16%

1,985.43 vs 2,360.48

fail → pass · 17fail → fail · 5

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-22 How do we write a system prompt template using Jinja2 syntax to format JSON outputs from a standard LLM completion call?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-03 We are starting a project to build an autonomous financial analyst agent. Which skills should be invoked during Phase 1: Agent Design?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-01 I am planning to build an autonomous AI coding assistant that can search internal code repositories, execute unit tests, and preserve context during long debugging sessions. Please provide an end-to-end development plan detailing the design architecture, tool integration strategy, memory implementation, and validation approach.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-21 How do we configure a Pinecone vector index for hybrid sparse-dense chunk retrieval using Qdrant or Pinecone SDKs without autonomous agent loops?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-04 We need to build a single standalone agent that handles customer support queries. Which skills should be invoked for the Single Agent Implementation phase?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-08 We need to set up short-term, long-term, and entity memory systems for our conversational agent. Which skills should be invoked for Phase 6?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-20 We want to perform LoRA fine-tuning on Llama 3 8B using PyTorch and Unsloth to improve its Python code generation syntax. How should we configure the hyperparameters and rank matrix?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-02 Our team has a basic LLM wrapper and we want to expand it into a full agent system that executes external API calls and maintains persistent long-term memory across user sessions. Please map out a step-by-step roadmap for tool interface design, dual-layer memory configuration, and quality assessment.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-05 Our engineering group is designing a multi-agent debate and research panel where agents take on distinct roles. Which specific framework and patterns skills should be invoked for Phase 3?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-19 Which related workflow bundles are associated with the AI Agent Development Workflow?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-07 We are defining custom API interfaces and schema definitions for our agent's action space. Which skills should be invoked during Phase 5: Tool Integration?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-17 In the general AI Agent Architecture model, describe the direct flow of components starting from User Input.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-10 Provide the exact copy-paste prompt syntax recommended for invoking architecture support in Phase 1 of agent development.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-16 Provide the exact copy-paste prompt recommended for measuring agent performance in Phase 7.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-11 Provide the exact copy-paste prompt recommended for implementing single agent patterns in Phase 2.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-18 List the five quality gate criteria that must be checked off before declaring an AI agent development workflow complete.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-06 We need stateful graph orchestration with persistence and conditional branching for our order processing agents. Which skills should be invoked during the Agent Orchestration phase?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-09 We are setting up automated evaluation and edge case testing for our deployed agent system. Which skills should be invoked during Phase 7?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-12 What is the recommended copy-paste prompt syntax for building role-based multi-agent systems in Phase 3?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-13 What copy-paste prompt should be used to create stateful workflows in Phase 4?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-14 What copy-paste prompt is recommended when building agent tools in Phase 5?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-15 Provide the exact copy-paste prompt recommended for memory system setup in Phase 6.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +77 percentage points is the difference between those two pass rates over the 21 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 0%
Pass rate with the skill: 77.3%
Cases attempted: 22
Cases counted toward lift: 21
Cases that regressed: 0
Skill version measured: v1
Run completed: 7/30/2026, 08:16 UTC

The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run77

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 59 days ago

Other measured skills in the registry, with their headline benchmark lift.