← Back to Skills Marketplace
152
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install agently-playbook
Description
Use when the user wants to build, initialize, validate, optimize, or refactor a model-powered assistant, internal tool, automation, evaluator, or workflow fr...
README (SKILL.md)
Agently Playbook
Use this skill first when the request still starts from business goals, refactor goals, product behavior, or broad model-app language.
The user does not need to say Agently, TriggerFlow, or any other framework term. Generic asks such as "build an assistant", "help me design an internal tool", or "create a validator for common problems" should still start here when the owner layer is unresolved.
Requests that also mention a UI, a web page, a desktop shell, or a local model service such as Ollama should still start here when the request is fundamentally about shaping a model-powered tool rather than only wiring one narrow capability.
Workflow
- Reduce the request into scenario and atomic goals.
- If the request is a project initialization or structure refactor, choose the owner layers, async boundary, and repo skeleton first.
- Choose the narrowest native Agently capability path.
- Name the concrete operations or primitives that should be used.
- Name the validation rule that proves the design stayed native-first.
Native-First Rules
- default to async-first guidance for service code, streaming, TriggerFlow, and any path that may overlap work or benefit from cancellation
- treat sync APIs as wrappers for scripts, REPL use, or compatibility bridges unless the host truly requires sync-only integration
- when the request is a project-shape refactor, separate settings, prompts, services, domain contracts, workflow, and tests before discussing low-level implementation details
Capability Routing
- model provider setup, settings-file-based model separation, or
${ENV.xxx}-backed settings loading ->agently-model-setup - request-side prompt design, prompt placeholder injection, or config-file prompt bridge ->
agently-prompt-management - output schema and reliability ->
agently-output-control - response reuse, metadata, or streaming consumption ->
agently-model-response - session continuity or restore ->
agently-session-memory - tools, MCP, FastAPIHelper,
auto_func, orKeyWaiter->agently-agent-extensions - embeddings, KB, or retrieval-to-answer ->
agently-knowledge-base - branching, concurrency, waiting/resume, mixed sync/async orchestration, event-driven fan-out, process-clarity refactors, runtime stream, or explicit multi-stage quality loops ->
agently-triggerflow - migration choice between LangChain and LangGraph ->
agently-migration-playbook
Anti-Patterns
- do not skip this playbook when the owner layer is unresolved
- do not invent custom output parsers, retry loops, or orchestration first
- do not let sync-first sample code dictate the service architecture when the target is clearly async-capable
- do not split project initialization into a fake standalone framework surface before the owner layers are chosen
- do not treat multi-agent, judge, or review flows as separate framework surfaces before checking native Agently capabilities
Read Next
references/capability-map.mdreferences/project-framework.md
Usage Guidance
This skill is a high-level design/playbook and appears coherent and low-risk by itself. Before using it in a production agent, remember: (1) it encourages using environment variables and .env patterns — do not paste secrets into chat or prompts; store provider keys in secure vaults or protected env variables; (2) follow-up work will be routed to specific agently-* leaf skills that may need provider credentials — review those leaf skills separately before granting secrets; (3) because it's instruction-only, it won't install code, but any code you scaffold based on its advice will. If you need the agent to work with live credentials, only provide them to trusted code/skills and verify those skills' install and env requirements first.
Capability Analysis
Type: OpenClaw Skill
Name: agently-playbook
Version: 0.1.0
The agently-playbook skill bundle consists entirely of architectural guidance and workflow instructions for an AI agent to assist users in building applications with the Agently framework. The files (SKILL.md, references/capability-map.md, and references/project-framework.md) define project structures, prompt management rules, and routing logic to other skills without containing any executable code, data exfiltration logic, or malicious prompt injection attempts.
Capability Assessment
Purpose & Capability
The name/description match the SKILL.md and reference files: this is a high-level playbook for shaping model-powered assistants and routing to leaf Agently capabilities. It does not request unrelated credentials, binaries, or system access.
Instruction Scope
The instructions are developer-facing guidance (project splitting, async-first rules, settings patterns) and do not direct the agent to read arbitrary user files or exfiltrate data. They do recommend using ${ENV.xxx} placeholders and validating env values at initialization — appropriate for a project playbook, but something to be aware of because downstream implementation steps may prompt for or require environment secrets.
Install Mechanism
No install spec or code files — instruction-only skill. Nothing is downloaded or written to disk by the skill itself.
Credentials
The skill declares no required environment variables or credentials. The references and SKILL.md recommend best-practice use of ${ENV.xxx} and .env patterns for implementations, which is proportionate to the stated purpose but means real implementations will later request provider keys (expected for downstream, implementation-specific skills).
Persistence & Privilege
always is false and there are no install hooks or config paths. The skill does not request permanent presence or system-level privileges.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install agently-playbook - After installation, invoke the skill by name or use
/agently-playbook - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
Initial release of agently-playbook.
- Introduces a skill for building, initializing, validating, optimizing, or refactoring model-powered assistants, internal tools, automations, or workflows from high-level business scenarios.
- Establishes a workflow to clarify goals, select owner layer, repo skeleton, and appropriate Agently capability path.
- Defines routing rules for specialized tasks like model setup, prompt management, output control, orchestration, and migrations.
- Emphasizes async-first patterns and discourages early low-level implementation or custom orchestration.
- Provides anti-patterns to avoid during early solution shaping.
- Lists relevant references for further guidance.
Metadata
Frequently Asked Questions
What is Agently Playbook?
Use when the user wants to build, initialize, validate, optimize, or refactor a model-powered assistant, internal tool, automation, evaluator, or workflow fr... It is an AI Agent Skill for Claude Code / OpenClaw, with 152 downloads so far.
How do I install Agently Playbook?
Run "/install agently-playbook" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Agently Playbook free?
Yes, Agently Playbook is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Agently Playbook support?
Agently Playbook is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Agently Playbook?
It is built and maintained by Maplemx (@maplemx); the current version is v0.1.0.
More Skills