Agents Best Practices
Install when you need a provider-neutral checklist to design, audit, or harden an agent harness (loops, tools, permissions, memory, MCP, evals) beyond a single coding-agent stack.
Overview
Agents Best Practices is a journey-wide agent skill that teaches provider-neutral agent harness architecture—usable whenever a solo builder needs to design or harden loops, tools, permissions, and observability before co
Install
npx skills add https://github.com/denissergeevitch/agents-best-practices --skill agents-best-practicesWhat is this skill?
- Provider-neutral harness model: instruction builder → model → tool proposal → schema validation → permission gate → exec
- Covers tool design, system prompts, planning/goals, context compaction, memory, skills, MCP, observability, evals, promp
- Treats coding agents as one subdomain; same patterns for research, ops, support, finance, and workflow automation agents
- Emphasizes harness as control plane: model proposes, runtime validates, authorizes, records, and summarizes
- Markdown-only file policy aligned with versioned agent-legible environments
- Default 8-step agent loop from user/task through context update and repeat
- Skill scope version 1.2.0 provider-neutral-agent-harness
Adoption & trust: 383 installs on skills.sh; 1.9k GitHub stars; 3/3 security scanners passed (skills.sh audits); trending (+100% hot-view momentum).
What problem does it solve?
You can call LLM APIs and wire tools, but without a rigorous control plane you get fragile loops, unsafe actions, opaque failures, and no repeatable way to evaluate harness changes.
Who is it for?
Solo builders shipping or refactoring custom agents (coding, research, ops, or vertical workflows) who want one neutral blueprint instead of vendor-specific blog scraps.
Skip if: Teams that only need a one-off prompt tweak with no tools, or builders who want a drop-in MCP server or hosted runtime rather than harness design principles.
When should I use this skill?
Designing, generating an MVP blueprint for, auditing, refactoring, or explaining an agentic harness for any domain.
What do I get? / Deliverables
You leave with a concrete default architecture and best-practice dimensions to implement or audit your harness, so model proposals stay validated, permissioned, logged, and improvable via evals and observability.
- Harness architecture aligned to the default control-plane loop
- Audit checklist across tools, memory, MCP, evals, and safety
- Refactor or MVP blueprint narrative for stakeholders
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Draft an MVP agent blueprint with explicit tool boundaries before building integrations.
Implement the default propose-validate-execute loop with structured observations for your domain agents.
Align permission decisions and approval pauses with safety requirements before enabling autonomous actions.
Define evals and feedback loops to regression-test harness changes across model providers.
Wire observability and compaction so long-running agents stay legible in production logs.
How it compares
Architecture and procedural doctrine for the agent runtime—not a single integration skill or a packaged data ETL template.
Common Questions / FAQ
Who is agents-best-practices for?
Indie and solo builders designing or debugging agent harnesses across domains (coding, research, support, ops) who need control-plane patterns independent of one API vendor.
When should I use agents-best-practices?
Use during Build when scaffolding agent-tooling; during Ship when adding evals and safety gates; during Operate when tightening observability and compaction; and during Validate when blueprinting an MVP agent before full implementation.
Is agents-best-practices safe to install?
It is markdown guidance with no built-in tool execution; review the Security Audits panel on this Prism page and treat permission and safety sections as design requirements in your own runtime.
SKILL.md
READMESKILL.md - Agents Best Practices
# Agents Best Practices Use this skill when the user asks how to build, improve, debug, or evaluate an agentic harness. This is a general-purpose agent architecture skill. Coding agents are one subdomain only; apply the same principles to research, finance, legal, support, operations, sales, healthcare, education, data analysis, procurement, and workflow automation agents. ## Core stance An agent harness is the control plane around a model. The model proposes actions; the harness validates, authorizes, executes, records, summarizes, and returns observations. Keep the loop simple and make the runtime rigorous. Default architecture: ```text user/task -> instruction and context builder -> model call -> tool/action proposal -> schema validation -> permission decision -> execution or approval pause -> structured observation -> context update -> repeat within budget or finish ``` ## When to activate this skill Use this skill for prompts involving any of these intents: - build an agent, agentic workflow, AI worker, autonomous assistant, or harness; - create a domain-specific MVP agent design, starter harness, implementation blueprint, or first production-safe version; - choose between OpenAI, Anthropic, OpenAI-compatible APIs, direct tool loops, hosted tools, or SDKs; - design tools, permissions, guardrails, approval flows, or sandboxing; - create planning mode, workflow orchestration, goal mode, todo tracking, or long-running task behavior; - add context compaction, memory, retrieval, scoped instructions, or prompt hierarchies; - attach Agent Skills, reusable workflows, MCP servers, external connectors, or tool search; - audit an existing agent for reliability, cost, prompt-cache hit rate, safety, latency, or observability; - create system prompts or developer instructions for a domain-specific agent; - make source-of-truth knowledge, validation signals, logs, metrics, or workflow state legible to an agent. Do not use this skill for ordinary single-turn writing, translation, or Q&A unless the user is asking about the design of an agent that will perform those tasks. ## How to use this skill First, identify the user's design problem: 1. **Domain**: what work the agent performs. 2. **Autonomy level**: answer-only, draft-only, approval-gated action, or autonomous action within policy. 3. **Risk level**: read-only, internal write, external communication, financial, legal, healthcare, security, destructive, or privileged. 4. **State duration**: single turn, multi-turn session, resumable workflow, or long-running goal. 5. **Tool surface**: internal APIs, hosted tools, MCP/external connectors, browser, sandbox, filesystem, database, communication, or computation. 6. **Validation**: what proves the task is complete. Then load the most relevant reference files, not all files by default. If the user asks to make or build an agent for a domain, default to MVP Builder Mode. ## MVP Builder Mode When the user asks to make, build, design, scaffold, or specify an agent for a domain, produce a concrete domain-specific MVP harness blueprint, not only advice. Use [mvp-agent-blueprint.md](references/mvp-agent-blueprint.md) as the primary reference and load other references as needed. Default behavior: 1. Infer a reasonable first version from the user's domain and stated constraints. 2. State assumptions briefly instead