Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/nibzard/awesome-agentic-patterns

A curated catalogue of awesome agentic AI patterns
https://github.com/nibzard/awesome-agentic-patterns

List: awesome-agentic-patterns

Last synced: 4 months ago
JSON representation

A curated catalogue of awesome agentic AI patterns

Awesome Lists containing this project

README 

          
# Awesome Agentic Patterns



![Awesome Agentic Patterns](/agentic-patterns.jpeg)



A curated catalogue of **agentic AI patterns** — real‑world tricks, workflows, and mini‑architectures that help autonomous or semi‑autonomous AI agents get useful work done in production.



> **Why?**

> Tutorials show toy demos. Real products hide the messy bits. This list surfaces the repeatable patterns that bridge the gap so we can all ship smarter, faster agents.



---



## What counts as a pattern?



* **Repeatable** – more than one team is using it.

* **Agent‑centric** – improves how an AI agent senses, reasons, or acts.

* **Traceable** – backed by a public reference: blog post, talk, repo, or paper.



If your link ticks those boxes, it belongs here.



---



## Quick Tour of Categories



|  Category                                              |  What you'll find                                         |

| ------------------------------------------------------ | --------------------------------------------------------- |

| [**Context & Memory**](#context-memory)                | Sliding‑window curation, vector cache, episodic memory    |

| [**Feedback Loops**](#feedback-loops)                  | Compilers, CI, human review, self‑healing retries         |

| [**Learning & Adaptation**](#learning-adaptation)      | Agent RFT, skill libraries, variance‑based RL             |

| [**Orchestration & Control**](#orchestration-control)  | Task decomposition, sub‑agent spawning, tool routing      |

| [**Reliability & Eval**](#reliability-eval)            | Guardrails, eval harnesses, logging, reproducibility      |

| [**Security & Safety**](#security-safety)              | Isolated VMs, PII tokenization, security scanning         |

| [**Tool Use & Environment**](#tool-use-environment)    | Shell, browser, DB, Playwright, sandbox tricks            |

| [**UX & Collaboration**](#ux-collaboration)            | Prompt hand‑offs, staged commits, async background agents |



*Categories are fluid — open a PR if you see a better slice!*

The tables below are auto‑generated from the `patterns/` folder.



---



### Context & Memory



- [Agent-Powered Codebase Q&A / Onboarding](patterns/agent-powered-codebase-qa-onboarding.md) UPDATED

- [Context Window Anxiety Management](patterns/context-window-anxiety-management.md) UPDATED

- [Context Window Auto-Compaction](patterns/context-window-auto-compaction.md)

- [Context-Minimization Pattern](patterns/context-minimization-pattern.md) UPDATED

- [Curated Code Context Window](patterns/curated-code-context-window.md) UPDATED

- [Curated File Context Window](patterns/curated-file-context-window.md) UPDATED

- [Dynamic Context Injection](patterns/dynamic-context-injection.md) UPDATED

- [Episodic Memory Retrieval & Injection](patterns/episodic-memory-retrieval-injection.md) UPDATED

- [Filesystem-Based Agent State](patterns/filesystem-based-agent-state.md) UPDATED

- [Layered Configuration Context](patterns/layered-configuration-context.md) UPDATED

- [Memory Synthesis from Execution Logs](patterns/memory-synthesis-from-execution-logs.md) UPDATED

- [Proactive Agent State Externalization](patterns/proactive-agent-state-externalization.md) UPDATED

- [Progressive Disclosure for Large Files](patterns/progressive-disclosure-large-files.md)

- [Prompt Caching via Exact Prefix Preservation](patterns/prompt-caching-via-exact-prefix-preservation.md)

- [Self-Identity Accumulation](patterns/self-identity-accumulation.md) UPDATED

- [Semantic Context Filtering Pattern](patterns/semantic-context-filtering.md)

- [Working Memory via TodoWrite](patterns/working-memory-via-todos.md)



### Feedback Loops



- [AI-Assisted Code Review / Verification](patterns/ai-assisted-code-review-verification.md) UPDATED

- [Background Agent with CI Feedback](patterns/background-agent-ci.md) UPDATED

- [Coding Agent CI Feedback Loop](patterns/coding-agent-ci-feedback-loop.md) UPDATED

- [Dogfooding with Rapid Iteration for Agent Improvement](patterns/dogfooding-with-rapid-iteration-for-agent-improvement.md) UPDATED

- [Graph of Thoughts (GoT)](patterns/graph-of-thoughts.md) UPDATED

- [Incident-to-Eval Synthesis](patterns/incident-to-eval-synthesis.md) NEW

- [Inference-Healed Code Review Reward](patterns/inference-healed-code-review-reward.md) UPDATED

- [Iterative Prompt & Skill Refinement](patterns/iterative-prompt-skill-refinement.md)

- [Reflection Loop](patterns/reflection.md) UPDATED

- [Rich Feedback Loops > Perfect Prompts](patterns/rich-feedback-loops.md) UPDATED

- [Self-Critique Evaluator Loop](patterns/self-critique-evaluator-loop.md) UPDATED

- [Self-Discover: LLM Self-Composed Reasoning Structures](patterns/self-discover-reasoning-structures.md) UPDATED

- [Spec-As-Test Feedback Loop](patterns/spec-as-test-feedback-loop.md) UPDATED

- [Tool Use Incentivization via Reward Shaping](patterns/tool-use-incentivization-via-reward-shaping.md) UPDATED



### Learning & Adaptation



- [Agent Reinforcement Fine-Tuning (Agent RFT)](patterns/agent-reinforcement-fine-tuning.md)

- [Compounding Engineering Pattern](patterns/compounding-engineering-pattern.md)

- [Frontier-Focused Development](patterns/frontier-focused-development.md)

- [Memory Reinforcement Learning (MemRL)](patterns/memory-reinforcement-learning-memrl.md)

- [Shipping as Research](patterns/shipping-as-research.md)

- [Skill Library Evolution](patterns/skill-library-evolution.md)

- [Variance-Based RL Sample Selection](patterns/variance-based-rl-sample-selection.md)



### Orchestration & Control



- [Action-Selector Pattern](patterns/action-selector-pattern.md) UPDATED

- [Agent Modes by Model Personality](patterns/agent-modes-by-model-personality.md)

- [Agent-Driven Research](patterns/agent-driven-research.md) UPDATED

- [Autonomous Workflow Agent Architecture](patterns/autonomous-workflow-agent-architecture.md)

- [Budget-Aware Model Routing with Hard Cost Caps](patterns/budget-aware-model-routing-with-hard-cost-caps.md) NEW

- [Burn the Boats](patterns/burn-the-boats.md)

- [Conditional Parallel Tool Execution](patterns/parallel-tool-execution.md) UPDATED

- [Continuous Autonomous Task Loop Pattern](patterns/continuous-autonomous-task-loop-pattern.md) UPDATED

- [Custom Sandboxed Background Agent](patterns/custom-sandboxed-background-agent.md)

- [Discrete Phase Separation](patterns/discrete-phase-separation.md)

- [Disposable Scaffolding Over Durable Features](patterns/disposable-scaffolding-over-durable-features.md) UPDATED

- [Distributed Execution with Cloud Workers](patterns/distributed-execution-cloud-workers.md)

- [Dual LLM Pattern](patterns/dual-llm-pattern.md) UPDATED

- [Explicit Posterior-Sampling Planner](patterns/explicit-posterior-sampling-planner.md) UPDATED

- [Factory over Assistant](patterns/factory-over-assistant.md)

- [Feature List as Immutable Contract](patterns/feature-list-as-immutable-contract.md)

- [Hybrid LLM/Code Workflow Coordinator](patterns/hybrid-llm-code-workflow-coordinator.md)

- [Inference-Time Scaling](patterns/inference-time-scaling.md) UPDATED

- [Initializer-Maintainer Dual Agent Architecture](patterns/initializer-maintainer-dual-agent.md)

- [Inversion of Control](patterns/inversion-of-control.md) UPDATED

- [Iterative Multi-Agent Brainstorming](patterns/iterative-multi-agent-brainstorming.md) UPDATED

- [Lane-Based Execution Queueing](patterns/lane-based-execution-queueing.md)

- [Language Agent Tree Search (LATS)](patterns/language-agent-tree-search-lats.md) UPDATED

- [LLM Map-Reduce Pattern](patterns/llm-map-reduce-pattern.md) UPDATED

- [Multi-Model Orchestration for Complex Edits](patterns/multi-model-orchestration-for-complex-edits.md) UPDATED

- [Opponent Processor / Multi-Agent Debate Pattern](patterns/opponent-processor-multi-agent-debate.md)

- [Oracle and Worker Multi-Model Approach](patterns/oracle-and-worker-multi-model.md) UPDATED

- [Parallel Tool Call Learning](patterns/parallel-tool-call-learning.md)

- [Plan-Then-Execute Pattern](patterns/plan-then-execute-pattern.md) UPDATED

- [Planner-Worker Separation for Long-Running Agents](patterns/planner-worker-separation-for-long-running-agents.md)

- [Progressive Autonomy with Model Evolution](patterns/progressive-autonomy-with-model-evolution.md) UPDATED

- [Progressive Complexity Escalation](patterns/progressive-complexity-escalation.md)

- [Recursive Best-of-N Delegation](patterns/recursive-best-of-n-delegation.md)

- [Self-Rewriting Meta-Prompt Loop](patterns/self-rewriting-meta-prompt-loop.md) UPDATED

- [Specification-Driven Agent Development](patterns/specification-driven-agent-development.md) UPDATED

- [Stop Hook Auto-Continue Pattern](patterns/stop-hook-auto-continue-pattern.md)

- [Sub-Agent Spawning](patterns/sub-agent-spawning.md)

- [Subject Hygiene for Task Delegation](patterns/subject-hygiene.md)

- [Swarm Migration Pattern](patterns/swarm-migration-pattern.md)

- [Three-Stage Perception Architecture](patterns/three-stage-perception-architecture.md) UPDATED

- [Tool Capability Compartmentalization](patterns/tool-capability-compartmentalization.md) UPDATED

- [Tool Selection Guide](patterns/tool-selection-guide.md)

- [Tree-of-Thought Reasoning](patterns/tree-of-thought-reasoning.md) UPDATED

- [Workspace-Native Multi-Agent Orchestration](patterns/workspace-native-multi-agent-orchestration.md) NEW



### Reliability & Eval



- [Action Caching & Replay Pattern](patterns/action-caching-replay.md)

- [Adaptive Sandbox Fan-Out Controller](patterns/adaptive-sandbox-fanout-controller.md)

- [Anti-Reward-Hacking Grader Design](patterns/anti-reward-hacking-grader-design.md)

- [Asynchronous Coding Agent Pipeline](patterns/asynchronous-coding-agent-pipeline.md) UPDATED

- [Canary Rollout and Automatic Rollback for Agent Policy Changes](patterns/canary-rollout-and-automatic-rollback-for-agent-policy-changes.md) NEW

- [CriticGPT-Style Code Review](patterns/criticgpt-style-evaluation.md) UPDATED

- [Extended Coherence Work Sessions](patterns/extended-coherence-work-sessions.md) UPDATED

- [Failover-Aware Model Fallback](patterns/failover-aware-model-fallback.md)

- [Lethal Trifecta Threat Model](patterns/lethal-trifecta-threat-model.md) UPDATED

- [LLM Observability](patterns/llm-observability.md)

- [Merged Code + Language Skill Model](patterns/merged-code-language-skill-model.md) UPDATED

- [No-Token-Limit Magic](patterns/no-token-limit-magic.md) UPDATED

- [Reliability Problem Map Checklist for RAG and Agents](patterns/wfgy-reliability-problem-map.md) NEW

- [RLAIF (Reinforcement Learning from AI Feedback)](patterns/rlaif-reinforcement-learning-from-ai-feedback.md) UPDATED

- [Schema Validation Retry with Cross-Step Learning](patterns/schema-validation-retry-cross-step-learning.md)

- [Structured Output Specification](patterns/structured-output-specification.md)

- [Versioned Constitution Governance](patterns/versioned-constitution-governance.md) UPDATED

- [Workflow Evals with Mocked Tools](patterns/workflow-evals-with-mocked-tools.md)



### Security & Safety



- [Deterministic Security Scanning Build Loop](patterns/deterministic-security-scanning-build-loop.md) UPDATED

- [External Credential Sync](patterns/external-credential-sync.md)

- [Hook-Based Safety Guard Rails for Autonomous Code Agents](patterns/hook-based-safety-guard-rails.md) UPDATED

- [Isolated VM per RL Rollout](patterns/isolated-vm-per-rl-rollout.md)

- [Non-Custodial Spending Controls](patterns/non-custodial-spending-controls.md) NEW

- [PII Tokenization](patterns/pii-tokenization.md) UPDATED

- [Sandboxed Tool Authorization](patterns/sandboxed-tool-authorization.md)

- [Soulbound Identity Verification](patterns/soulbound-identity-verification.md) NEW

- [Zero-Trust Agent Mesh](patterns/zero-trust-agent-mesh.md) NEW



### Tool Use & Environment



- [Agent SDK for Programmatic Control](patterns/agent-sdk-for-programmatic-control.md) UPDATED

- [Agent-First Tooling and Logging](patterns/agent-first-tooling-and-logging.md) UPDATED

- [Agentic Search Over Vector Embeddings](patterns/agentic-search-over-vector-embeddings.md)

- [AI Web Search Agent Loop](patterns/ai-web-search-agent-loop.md)

- [CLI-First Skill Design](patterns/cli-first-skill-design.md) UPDATED

- [CLI-Native Agent Orchestration](patterns/cli-native-agent-orchestration.md) UPDATED

- [Code Mode MCP Tool Interface Improvement Pattern](patterns/code-first-tool-interface-pattern.md) UPDATED

- [Code-Over-API Pattern](patterns/code-over-api-pattern.md) UPDATED

- [Code-Then-Execute Pattern](patterns/code-then-execute-pattern.md) UPDATED

- [Dual-Use Tool Design](patterns/dual-use-tool-design.md)

- [Dynamic Code Injection (On-Demand File Fetch)](patterns/dynamic-code-injection-on-demand-file-fetch.md) UPDATED

- [Egress Lockdown (No-Exfiltration Channel)](patterns/egress-lockdown-no-exfiltration-channel.md) UPDATED

- [Intelligent Bash Tool Execution](patterns/intelligent-bash-tool-execution.md)

- [LLM-Friendly API Design](patterns/llm-friendly-api-design.md) UPDATED

- [Multi-Platform Communication Aggregation](patterns/multi-platform-communication-aggregation.md) UPDATED

- [Multi-Platform Webhook Triggers](patterns/multi-platform-webhook-triggers.md)

- [Patch Steering via Prompted Tool Selection](patterns/patch-steering-via-prompted-tool-selection.md) UPDATED

- [Progressive Tool Discovery](patterns/progressive-tool-discovery.md) UPDATED

- [Shell Command Contextualization](patterns/shell-command-contextualization.md) UPDATED

- [Subagent Compilation Checker](patterns/subagent-compilation-checker.md) UPDATED

- [Tool Use Steering via Prompting](patterns/tool-use-steering-via-prompting.md) UPDATED

- [Virtual Machine Operator Agent](patterns/virtual-machine-operator-agent.md) UPDATED

- [Visual AI Multimodal Integration](patterns/visual-ai-multimodal-integration.md) UPDATED



### UX & Collaboration



- [Abstracted Code Representation for Review](patterns/abstracted-code-representation-for-review.md) UPDATED

- [Agent-Assisted Scaffolding](patterns/agent-assisted-scaffolding.md) UPDATED

- [Agent-Friendly Workflow Design](patterns/agent-friendly-workflow-design.md) UPDATED

- [AI-Accelerated Learning and Skill Development](patterns/ai-accelerated-learning-and-skill-development.md) UPDATED

- [Chain-of-Thought Monitoring & Interruption](patterns/chain-of-thought-monitoring-interruption.md)

- [Codebase Optimization for Agents](patterns/codebase-optimization-for-agents.md)

- [Democratization of Tooling via Agents](patterns/democratization-of-tooling-via-agents.md) UPDATED

- [Dev Tooling Assumptions Reset](patterns/dev-tooling-assumptions-reset.md)

- [Human-in-the-Loop Approval Framework](patterns/human-in-loop-approval-framework.md)

- [Latent Demand Product Discovery](patterns/latent-demand-product-discovery.md)

- [Milestone Escrow for Agent Resource Funding](patterns/agentfund-crowdfunding.md) NEW

- [Proactive Trigger Vocabulary](patterns/proactive-trigger-vocabulary.md) UPDATED

- [Seamless Background-to-Foreground Handoff](patterns/seamless-background-to-foreground-handoff.md) UPDATED

- [Spectrum of Control / Blended Initiative](patterns/spectrum-of-control-blended-initiative.md) UPDATED

- [Team-Shared Agent Configuration as Code](patterns/team-shared-agent-configuration.md)

- [Verbose Reasoning Transparency](patterns/verbose-reasoning-transparency.md) UPDATED



---



## Contributing in 3 steps



1. **Fork & branch** → `git checkout -b add-my-pattern`

2. **Add a pattern file** under `patterns/` using the template above.

3. **Open a PR** titled `Add: my-pattern-name` — the README & site will regenerate automatically.

4. This repository is pattern-first: proposals that are primarily product announcements or promotions will be rejected, even if technically valid.



See [`CONTRIBUTING.md`](https://github.com/nibzard/awesome-agentic-patterns/blob/main/CONTRIBUTING.md) for the fine print.



---



## Inspiration



This project started after the write‑up [**"What Sourcegraph learned building AI coding agents"**](https://www.nibzard.com/ampcode) (28 May 2025) and the ongoing *Raising an Agent* video diary. Many first patterns come straight from those lessons — thanks to everyone sharing their journey in the open!



---



## License



Apache‑2.0. See [`LICENSE`](https://github.com/nibzard/awesome-agentic-patterns/blob/main/LICENSE).



---



## Star History



[![Star History Chart](https://api.star-history.com/svg?repos=nibzard/awesome-agentic-patterns&type=date&legend=top-left)](https://www.star-history.com/#nibzard/awesome-agentic-patterns&type=date&legend=top-left)