AI Emotional Companion Chat
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md and install @user_2cc194fa/skill-hgl.
About this skill
Problem It Solves
When an app needs continuous, context-aware conversation but does not want to build a full chat front end, session state machine, and service proxy, this skill wraps a third-party Agent API into a triggerable chat agent capability. It targets explicit intents such as 'invoke a chatbot / emotional companion chat' and focuses on the multi-turn problem of remembering prior context, especially when users want to continue a previous session with session_id.
How It Works and Boundaries
The skill depends on an external Agent service and requires api_url, agent_id, and api_key. At runtime it sends the user's message as query and can optionally include session_id; if omitted, a new conversation is started. It calls POST {api_url}/api/v1/agents/{agent_id}/completions with Bearer Token authentication and a 30 second timeout. The response contains answer for the Agent reply, session_id for continuation, and error when the request fails.
- Core capability: multi-turn dialogue, context continuation, and a consistent response shape
- Key steps: read configuration → send the Agent API request → parse the reply and session ID
- Boundary: depends on remote service availability, requires careful API key handling, and is not suitable for offline use, strict local data isolation, or complex tool orchestration
Use Cases
- When handling post-ticket user soothing, call the skill to generate continuous companion replies and continue context via session_id.
- During integration tests for a remote Agent API, use the skill to verify query, session_id, and answer response fields.
- When building a chat prototype, wrap the third-party Agent service into a continuable multi-turn interface.
- While debugging the 30-second timeout and Bearer auth flow, send a fixed query to check reply and error output.
Best For
- Engineers building customer-support tone prototypes: need quick remote Agent access and multi-turn reply validation.
- Front-end engineers building chat demos: need to wrap API calls into session_id-continuable dialogue endpoints.
- QA engineers testing third-party Agent services: need to verify answer, session_id, error fields, and timeout behavior.
- Independent developers building AI companion assistants: want to avoid custom session management and call existing Agent APIs.
Related Skills
Turn vague requests into structured prompts with role, task, context, and constraints, outputting only usable prompts without executing the task.
A skill for game NPC and enemy AI that separates decision, steering, and pathfinding, covering FSMs, behavior trees, A*, path caching, and pitfalls.
An evaluation-driven workflow for diagnosing, optimizing, validating, and logging Agent Skills based on design patterns and anti-patterns.
Analyzes web form structure in a headed browser, fills choice-based answers, verifies them against a docx file, and leaves personal fields for manual input.