用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/diegosouzapw/awesome-omni-skill --skill jules命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | jules |
| description | Integration with Google's Jules API for agentic coding sessions. |
| license | UNLICENSED |
| metadata | {"author":"user","version":"1.0.0"} |
This skill provides instructions for interacting with the Google Jules API. Use this skill when you need to create coding sessions, manage tasks, or interact with the Jules agent.
/home/cgasgarth/.config/opencode/jules-secret.json.
{ "julesApiKey": "YOUR_KEY" }
X-Goog-Api-Key: YOUR_KEY in all requests.https://jules.googleapis.com/v1alpha
GET /sources
Returns available sources (e.g., GitHub repositories) connected to the user.
POST /sessions
Creates a new coding session.
Body:
{
"prompt": "Task description",
"sourceContext": {
"source": "projects/PROJECT_ID/locations/global/connections/CONNECTION_ID/repositories/REPO_ID",
"githubRepoContext": {
"startingBranch": "main"
}
},
"automationMode": "AUTO_CREATE_PR", // Optional
"requirePlanApproval": true // Optional, defaults to false
}
GET /sessions
Lists active and past sessions.
GET /sessions/{sessionId}/activities
Returns the history of activities in a session.
Parameters:
pageSize (optional): Number of activities to return (e.g., ?pageSize=30).Response Structure:
Returns an activities array. Activity types include:
planGenerated: Agent created a plan.planApproved: User approved a plan.progressUpdated: Agent progress message (contains title, description).sessionCompleted: Session finished.Artifacts:
Activities may contain an artifacts field with data like bash output, git patches, or images.
POST /sessions/{sessionId}:sendMessage
Sends a user message to the session.
Body:
{
"prompt": "Your message here"
}
POST /sessions/{sessionId}:approvePlan
Approves the plan proposed by Jules to proceed with implementation.
To list sessions (replace KEY with actual key):
curl "https://jules.googleapis.com/v1alpha/sessions" \
-H "X-Goog-Api-Key: $KEY"
To get session activities:
curl "https://jules.googleapis.com/v1alpha/sessions/$SESSION_ID/activities?pageSize=30" \
-H "X-Goog-Api-Key: $KEY"
To create a session:
curl -X POST "https://jules.googleapis.com/v1alpha/sessions" \
-H "X-Goog-Api-Key: $KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "Fix the bug in login",
"sourceContext": { ... }
}'
Token-efficient tracking for AI orchestration. CLI-first for status updates (~50 tokens), agent fallback for complex ops (~1KB). Use when: updating task status, querying blockers, creating progress files, validating phases.
AshAi extension guidelines for integrating AI capabilities with Ash Framework. Use when implementing vectorization/embeddings, exposing Ash actions as LLM tools, creating prompt-backed actions, or setting up MCP servers. Covers semantic search, LangChain integration, and structured outputs.
This skill should be used when solving hard questions, complex architectural problems, or debugging issues that benefit from GPT-5 Pro or GPT-5.1 thinking models with large file context. Use when standard Claude analysis needs deeper reasoning or extended context windows.
基于 SOC 职业分类