agents-best-practices
bydenissergeevitch· 1 skill
Use this skill when designing, generating an MVP blueprint for, auditing, refactoring, or explaining an agentic harness for any domain. Covers provider-neutral agent architecture for OpenAI, Anthropic, and OpenAI-compatible APIs: agent loops, tool design, permissions, system prompts, planning, goals, context compaction, memory, skills, MCP/external connectors, observability, evals, prompt caching, agent-legible environments, feedback loops, and safety.
How to install agents-best-practices?
denissergeevitch/agents-best-practices
$npx -y skills add denissergeevitch/agents-best-practices --skill agents-best-practicesInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Run `npx skills use "https://github.com/denissergeevitch/agents-best-practices" --skill "denissergeevitch/agents-best-practices"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/denissergeevitch/agents-best-practices" that are relevant to the current task. Run `npx skills add "https://github.com/denissergeevitch/agents-best-practices"` and select the relevant skills, then follow their instructions.
SKILL.md
| 1 | <div align="center"> |
| 2 | |
| 3 | # agents-best-practices |
| 4 | |
| 5 | <img src="icon.jpeg" alt="agents-best-practices icon" width="300" /> |
| 6 | |
| 7 | > *"The model proposes actions; the harness validates, authorizes, executes, records, and returns observations."* |
| 8 | |
| 9 | [](LICENSE) |
| 10 | [](SKILL.md) |
| 11 | [](SKILL.md) |
| 12 | [](SKILL.md) |
| 13 | |
| 14 | <br> |
| 15 | |
| 16 | **A provider-neutral Agent Skill for designing, generating MVP blueprints for, auditing, refactoring, and explaining agentic harnesses.** |
| 17 | |
| 18 | It applies beyond coding agents: research, support, operations, sales, finance, data analysis, procurement, legal workflows, healthcare workflows, education, and workflow automation agents all need the same core runtime discipline. |
| 19 | |
| 20 | <br> |
| 21 | |
| 22 | **Install** - pick one: |
| 23 | |
| 24 | </div> |
| 25 | |
| 26 | **A. With [`skills`](https://github.com/vercel-labs/skills) (any compatible agent):** |
| 27 | |
| 28 | ```bash |
| 29 | npx skills add DenisSergeevitch/agents-best-practices -g |
| 30 | ``` |
| 31 | |
| 32 | The `-g` flag installs globally at user level so every project can discover it. |
| 33 | |
| 34 | **B. Or paste this prompt to your AI agent:** |
| 35 | |
| 36 | ```text |
| 37 | Install the agents-best-practices skill for me: |
| 38 | |
| 39 | 1. Clone https://github.com/DenisSergeevitch/agents-best-practices into my |
| 40 | user-level skills directory as `agents-best-practices/`. |
| 41 | Use the skill directory my agent reads on this machine, for example: |
| 42 | - Codex: ~/.codex/skills/ |
| 43 | - Claude Code: ~/.claude/skills/ |
| 44 | 2. Verify that SKILL.md, icon.jpeg, and the references/ directory are present. |
| 45 | 3. Confirm the install path when done. |
| 46 | ``` |
| 47 | |
| 48 | **C. Manual install paths:** |
| 49 | |
| 50 | ```bash |
| 51 | # Codex |
| 52 | mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills" |
| 53 | git clone https://github.com/DenisSergeevitch/agents-best-practices.git \ |
| 54 | "${CODEX_HOME:-$HOME/.codex}/skills/agents-best-practices" |
| 55 | |
| 56 | # Claude Code, user-level |
| 57 | mkdir -p "$HOME/.claude/skills" |
| 58 | git clone https://github.com/DenisSergeevitch/agents-best-practices.git \ |
| 59 | "$HOME/.claude/skills/agents-best-practices" |
| 60 | |
| 61 | # Claude Code, project-level |
| 62 | mkdir -p .claude/skills |
| 63 | git clone https://github.com/DenisSergeevitch/agents-best-practices.git \ |
| 64 | .claude/skills/agents-best-practices |
| 65 | ``` |
| 66 | |
| 67 | <div align="center"> |
| 68 | |
| 69 | [Use cases](#use-cases) · [What this is](#what-this-is) · [Philosophy](#philosophy) · [Layout](#layout) · [Sources](#sources) |
| 70 | |
| 71 | </div> |
| 72 | |
| 73 | --- |
| 74 | |
| 75 | ## Use cases |
| 76 | |
| 77 | <p align="center"> |
| 78 | <img src="assets/agents-best-practices-illustrations/01-use-cases.jpg" alt="Xiaohei sorting agent harness use cases into MVP blueprint, audit fixes, and permission map" width="760" /> |
| 79 | </p> |
| 80 | |
| 81 | Once installed, the skill activates when a conversation touches agent architecture, harness design, tool permissions, planning mode, workflow orchestration, context and memory, skills, connectors, observability, evals, prompt caching, or production readiness. Three common ways to use it: |
| 82 | |
| 83 | ### Case 1 - Generate an MVP agent blueprint |
| 84 | |
| 85 | You have a domain and need the smallest useful production-safe agent harness, not a vague list of best practices. |
| 86 | |
| 87 | ```text |
| 88 | You > Build an agent for account renewal risk. It should read CRM, |
| 89 | support tickets, and usage data, then draft renewal actions. |
| 90 | |
| 91 | Agent > Start with an approval-gated Level 2 harness. |
| 92 | The MVP has one job: produce a renewal-risk brief plus draft |
| 93 | next actions for a human account owner. |
| 94 | |
| 95 | Core loop: |
| 96 | user/task -> context builder -> model call -> typed tool call |
| 97 | -> schema validation -> permission check -> execution or pause |
| 98 | -> structured observation -> next step or final brief |
| 99 | |
| 100 | Minimal tools: |
| 101 | - read_account_profile read_private_data |
| 102 | - list_support_tickets read_private_data |
| 103 | - fetch_usage_summary read_private_data |
| 104 | - draft_customer_email draft_external_message |
| 105 | - request_approval approval_gate |
| 106 | |
| 107 | L |
Related
6 picksType
- find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express…Jul 2026
- mcp-builderGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools.Jul 2026
- google-agents-cli-adk-codeThis skill should be used when the user wants to "write agent code", "build an agent with ADK", "add a tool", "create a callback", "define an agent", "use…Jul 2026
- google-agents-cli-workflowThis skill should be used when the user wants to "develop an agent", "build an agent using ADK", "run the agent locally", "debug agent code", "test an agent",…Jul 2026
- google-agents-cli-evalThis skill should be used when the user wants to "run an evaluation", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval…Jul 2026
- google-agents-cli-scaffoldThis skill should be used when the user wants to "create an agent project", "start a new ADK project", "build me a new agent", "add CI/CD to my project", "add…Jul 2026
