[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CTCD7SEXCFA53XNPYVT8RZ":3,"skill-same:01M7CTCD7SEXCFA53XNPYVT8RZ":44},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M7CTCD7SEXCFA53XNPYVT8RZ","build-omnigent",null,"Patterns and templates for generating valid Omnigent agent directories. Load when ready to create files.","roland-id-au\u002Fomniagent","roland-id-au","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F2590653?v=4","01M7CT4DEBVK4R185WFWCWN4JD","Buck Chancey","01M7CT4DFGWX49CGE347EF4V6T",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Fomnigent\u002Fonboarding\u002Fagent\u002Fskills\u002Fbuild-omnigent\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: build-omnigent\ndescription: Patterns and templates for generating valid Omnigent agent directories. Load when ready to create files.\n---\n\n# Agent Generation\n\nUse these patterns to generate a valid agent directory. Always generate\nthe minimal set of files needed — don't over-engineer.\n\nEvery template below has been validated with the same parser\u002Fvalidator\nthat `omnigent server` uses. If your environment exposes the\n`validate_agent` tool (the dedicated agent-authoring environment does),\nrun it after generating files to confirm the spec loads. Load the\n**`omnigent-knowledge`** skill if you need the deeper field reference\n(executor types, os_env, guardrails, sandboxing).\n\n## Step 1: Choose a directory name\n\nUse the agent name in kebab-case: `my-research-agent\u002F`\n\n## Step 2: Generate config.yaml\n\nAlways include:\n- `spec_version: 1`\n- `name` (lowercase, hyphens OK)\n- `description` (one sentence)\n- `instructions` — path to a file (default `AGENTS.md`) or inline text.\n  (`prompt:` is an accepted alias; `instructions:` wins if both are set.)\n- `executor` — how the agent runs. See Step 2a.\n\nInclude if needed:\n- `tools.builtins` — built-in tools. The current set is `download_file`,\n  `export_agent`, `list_files`, `search_conversations`, `upload_file`,\n  `web_fetch`, `web_search`. If the `list_builtin_tools` tool is available,\n  call it for the authoritative live set rather than trusting this list.\n- `tools.agents` — sub-agents, by the `name` each declares under `agents\u002F`\n  (a sub-agent's directory name may differ from its name).\n- `os_env` — filesystem\u002Fshell access for harness agents (see the\n  shell-capable template).\n- `interaction.modalities` — if the agent handles images or files.\n- `guardrails` — runtime policy gates (see `omnigent-knowledge`).\n\n## Step 2a: Choose an executor\n\n`executor.type` must be one of **`claude_sdk`**, **`agents_sdk`**, or\n**`omnigent`**. There is **no `llm` executor** — do not use it.\n\n| Need | executor |\n|------|----------|\n| A fresh, simple LLM agent (default) | `claude_sdk` (Anthropic) or `agents_sdk` (OpenAI), in-process |\n| Existing Claude SDK \u002F OpenAI Agents SDK code | `claude_sdk` \u002F `agents_sdk` |\n| A CLI\u002Fcoding harness, shell + file tools, sub-agents | `omnigent` + a `config.harness` |\n\nWhen `executor.type: omnigent`, **`config.harness` is required** and must\nbe one of: `claude-native` (Claude Code, full coding tools), `claude-sdk`,\n`codex-native`, `codex`, `openai-agents`, `open-responses`, `pi`.\n(`claude` is an alias for `claude-native`.)\n\nModel selection is optional — if omitted, the executor resolves the\nprovider's default model from the configured credentials (e.g. an\nAnthropic key, a Claude subscription, or a Databricks profile). Pin one\nonly when asked; see `omnigent-knowledge` for `executor.model` \u002F `auth`.\n\n## Step 3: Generate AGENTS.md\n\nWrite a focused system prompt:\n- Identity: \"You are a [role] that [does what].\"\n- Capabilities: what tools\u002Fskills are available\n- Constraints: what NOT to do\n- Style: how to communicate\n\nKeep it under 500 words for a starter agent. The user can expand later.\n\n## Step 4: Generate skills (optional)\n\nOnly generate skills if the agent has distinct modes of operation.\nEach skill needs:\n\n```\nskills\u002F\u003Cdir>\u002FSKILL.md\n```\n\nThe directory name is free-form and need not match the skill's `name` —\nthe runtime identifies a skill by its frontmatter `name` and loads its files\nfrom whatever directory it sits in.\n\nWith YAML frontmatter:\n```markdown\n---\nname: skill-name\ndescription: One-line description of what this skill does.\n---\n\nDetailed instructions for when this skill is loaded...\n```\n\n## Templates\n\n### Minimal agent (simplest — in-process SDK)\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: claude_sdk      # or agents_sdk for OpenAI\ninstructions: AGENTS.md\n```\n\n**AGENTS.md:**\n```markdown\nYou are {agent_name}, {description}.\n\nAnswer questions clearly and concisely. If you don't know something,\nsay so rather than guessing.\n```\n\n### Agent with web search\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: claude_sdk\ntools:\n  builtins:\n    - web_search        # one of the builtins listed in Step 2\ninteraction:\n  modalities:\n    input: [text]\n    output: [text]\ninstructions: AGENTS.md\n```\n\n### Harness agent with shell + filesystem access\n\nUse the `omnigent` executor with a coding harness when the agent needs to\nrun commands and read\u002Fwrite files. `os_env` grants OS access; the harness\nexposes `sys_os_read` \u002F `sys_os_write` \u002F `sys_os_edit` \u002F `sys_os_shell`.\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: omnigent\n  config:\n    harness: claude-native\n    # Headless runs can't answer approval prompts — bypass them. Pair\n    # with a read-only prompt and\u002For a blast_radius guardrail for safety.\n    permission_mode: bypassPermissions   # codex-native uses `yolo: true`\nos_env:\n  type: caller_process\n  cwd: .\n  sandbox:\n    type: none          # or linux_bwrap \u002F darwin_seatbelt to sandbox\ninstructions: AGENTS.md\n```\n\n### Agent with MCP server integration\n\n**Directory structure:**\n```\n{agent_name}\u002F\n  config.yaml\n  AGENTS.md\n  tools\u002F\n    mcp\u002F\n      github.yaml\n```\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: claude_sdk\ninstructions: AGENTS.md\n```\n\n**tools\u002Fmcp\u002Fgithub.yaml:**\n```yaml\ntransport: http\nurl: https:\u002F\u002Fyour-mcp-server.example.com\u002Fsse\nheaders:\n  Authorization: Bearer ${{{mcp_token_var}}}\n```\n\n### Multi-agent system with sub-agents\n\nThe **parent** needs the `omnigent` executor — that's what provides the spawn\ntools. Each sub-agent is a full agent and may use any executor.\n\n**Directory structure:**\n```\n{agent_name}\u002F\n  config.yaml\n  AGENTS.md\n  agents\u002F\n    {sub_agent_1_dir}\u002F\n      config.yaml\n    {sub_agent_2_dir}\u002F\n      config.yaml\n```\n\nDirectory names are free-form. A sub-agent's identity is the `name` in its\nown `config.yaml`, and that is what the parent lists in `tools.agents` —\n`{sub_agent_1_dir}` and `{sub_agent_1}` may differ.\n\n**Parent config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: omnigent\n  config:\n    harness: claude-sdk\ntools:\n  agents:\n    - {sub_agent_1}\n    - {sub_agent_2}\ninstructions: AGENTS.md\n```\n\n**Sub-agent config (agents\u002F{sub_agent_1_dir}\u002Fconfig.yaml):**\n```yaml\nspec_version: 1\nname: {sub_agent_1}\ndescription: {sub_agent_1_description}\nexecutor:            # any executor works here — only the parent needs omnigent\n  type: omnigent\n  config:\n    harness: claude-sdk\ninstructions: |\n  You are {sub_agent_1}. {sub_agent_1_instructions}\n```\n\n**Parent AGENTS.md should reference sub-agents:**\n```markdown\nYou have sub-agents you can delegate to:\n- **{sub_agent_1}** — {sub_agent_1_description}\n- **{sub_agent_2}** — {sub_agent_2_description}\n\nCall `sys_session_send(type=\"\u003Cname>\", input=\"\u003Ctask>\")` to dispatch a\ndeclared sub-agent. Emit multiple `sys_session_send` tool calls in the\nsame response to run them in parallel; results arrive via the inbox.\n```\n\n## Environment variable naming conventions\n\nWhen pinning credentials with `${ENV_VAR}`, map providers to their\nstandard env var names:\n- `openai` → `OPENAI_API_KEY`\n- `anthropic` → `ANTHROPIC_API_KEY`\n- `gemini` → `GEMINI_API_KEY` or `GOOGLE_API_KEY`\n- `groq` → `GROQ_API_KEY`\n- `deepseek` → `DEEPSEEK_API_KEY`\n- `xai` → `XAI_API_KEY`\n- `mistral` → `MISTRAL_API_KEY`\n- `databricks` → `DATABRICKS_TOKEN` (or an `auth.profile`)\n\n## Validation checklist\n\nBefore presenting the generated files to the user, verify (and if\n`validate_agent` is available, run it to confirm):\n- [ ] `spec_version: 1` is present\n- [ ] `name` is set and uses lowercase + hyphens\n- [ ] `executor.type` is one of `claude_sdk`, `agents_sdk`, `omnigent`\n- [ ] If `executor.type: omnigent`, `executor.config.harness` is set to a\n      valid harness\n- [ ] `instructions` (or `prompt`) points to a file that exists or is\n      inline text\n- [ ] When declaring `tools.agents`, the parent uses `executor.type:\n      omnigent`, and each entry is the declared `name` of a sub-agent under\n      `agents\u002F` (its directory name may differ; sub-agents may use any\n      executor)\n- [ ] `tools.builtins` names are from the known set (Step 2) — or, if\n      `list_builtin_tools` is available, were confirmed against it\n- [ ] Skill names use the `[a-z0-9-]+` pattern (they need not match their\n      directory names)\n","# Agent Generation\n\nUse these patterns to generate a valid agent directory. Always generate\nthe minimal set of files needed — don't over-engineer.\n\nEvery template below has been validated with the same parser\u002Fvalidator\nthat `omnigent server` uses. If your environment exposes the\n`validate_agent` tool (the dedicated agent-authoring environment does),\nrun it after generating files to confirm the spec loads. Load the\n**`omnigent-knowledge`** skill if you need the deeper field reference\n(executor types, os_env, guardrails, sandboxing).\n\n## Step 1: Choose a directory name\n\nUse the agent name in kebab-case: `my-research-agent\u002F`\n\n## Step 2: Generate config.yaml\n\nAlways include:\n- `spec_version: 1`\n- `name` (lowercase, hyphens OK)\n- `description` (one sentence)\n- `instructions` — path to a file (default `AGENTS.md`) or inline text.\n  (`prompt:` is an accepted alias; `instructions:` wins if both are set.)\n- `executor` — how the agent runs. See Step 2a.\n\nInclude if needed:\n- `tools.builtins` — built-in tools. The current set is `download_file`,\n  `export_agent`, `list_files`, `search_conversations`, `upload_file`,\n  `web_fetch`, `web_search`. If the `list_builtin_tools` tool is available,\n  call it for the authoritative live set rather than trusting this list.\n- `tools.agents` — sub-agents, by the `name` each declares under `agents\u002F`\n  (a sub-agent's directory name may differ from its name).\n- `os_env` — filesystem\u002Fshell access for harness agents (see the\n  shell-capable template).\n- `interaction.modalities` — if the agent handles images or files.\n- `guardrails` — runtime policy gates (see `omnigent-knowledge`).\n\n## Step 2a: Choose an executor\n\n`executor.type` must be one of **`claude_sdk`**, **`agents_sdk`**, or\n**`omnigent`**. There is **no `llm` executor** — do not use it.\n\n| Need | executor |\n|------|----------|\n| A fresh, simple LLM agent (default) | `claude_sdk` (Anthropic) or `agents_sdk` (OpenAI), in-process |\n| Existing Claude SDK \u002F OpenAI Agents SDK code | `claude_sdk` \u002F `agents_sdk` |\n| A CLI\u002Fcoding harness, shell + file tools, sub-agents | `omnigent` + a `config.harness` |\n\nWhen `executor.type: omnigent`, **`config.harness` is required** and must\nbe one of: `claude-native` (Claude Code, full coding tools), `claude-sdk`,\n`codex-native`, `codex`, `openai-agents`, `open-responses`, `pi`.\n(`claude` is an alias for `claude-native`.)\n\nModel selection is optional — if omitted, the executor resolves the\nprovider's default model from the configured credentials (e.g. an\nAnthropic key, a Claude subscription, or a Databricks profile). Pin one\nonly when asked; see `omnigent-knowledge` for `executor.model` \u002F `auth`.\n\n## Step 3: Generate AGENTS.md\n\nWrite a focused system prompt:\n- Identity: \"You are a [role] that [does what].\"\n- Capabilities: what tools\u002Fskills are available\n- Constraints: what NOT to do\n- Style: how to communicate\n\nKeep it under 500 words for a starter agent. The user can expand later.\n\n## Step 4: Generate skills (optional)\n\nOnly generate skills if the agent has distinct modes of operation.\nEach skill needs:\n\n```\nskills\u002F\u003Cdir>\u002FSKILL.md\n```\n\nThe directory name is free-form and need not match the skill's `name` —\nthe runtime identifies a skill by its frontmatter `name` and loads its files\nfrom whatever directory it sits in.\n\nWith YAML frontmatter:\n```markdown\n---\nname: skill-name\ndescription: One-line description of what this skill does.\n---\n\nDetailed instructions for when this skill is loaded...\n```\n\n## Templates\n\n### Minimal agent (simplest — in-process SDK)\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: claude_sdk      # or agents_sdk for OpenAI\ninstructions: AGENTS.md\n```\n\n**AGENTS.md:**\n```markdown\nYou are {agent_name}, {description}.\n\nAnswer questions clearly and concisely. If you don't know something,\nsay so rather than guessing.\n```\n\n### Agent with web search\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: claude_sdk\ntools:\n  builtins:\n    - web_search        # one of the builtins listed in Step 2\ninteraction:\n  modalities:\n    input: [text]\n    output: [text]\ninstructions: AGENTS.md\n```\n\n### Harness agent with shell + filesystem access\n\nUse the `omnigent` executor with a coding harness when the agent needs to\nrun commands and read\u002Fwrite files. `os_env` grants OS access; the harness\nexposes `sys_os_read` \u002F `sys_os_write` \u002F `sys_os_edit` \u002F `sys_os_shell`.\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: omnigent\n  config:\n    harness: claude-native\n    # Headless runs can't answer approval prompts — bypass them. Pair\n    # with a read-only prompt and\u002For a blast_radius guardrail for safety.\n    permission_mode: bypassPermissions   # codex-native uses `yolo: true`\nos_env:\n  type: caller_process\n  cwd: .\n  sandbox:\n    type: none          # or linux_bwrap \u002F darwin_seatbelt to sandbox\ninstructions: AGENTS.md\n```\n\n### Agent with MCP server integration\n\n**Directory structure:**\n```\n{agent_name}\u002F\n  config.yaml\n  AGENTS.md\n  tools\u002F\n    mcp\u002F\n      github.yaml\n```\n\n**config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: claude_sdk\ninstructions: AGENTS.md\n```\n\n**tools\u002Fmcp\u002Fgithub.yaml:**\n```yaml\ntransport: http\nurl: https:\u002F\u002Fyour-mcp-server.example.com\u002Fsse\nheaders:\n  Authorization: Bearer ${{{mcp_token_var}}}\n```\n\n### Multi-agent system with sub-agents\n\nThe **parent** needs the `omnigent` executor — that's what provides the spawn\ntools. Each sub-agent is a full agent and may use any executor.\n\n**Directory structure:**\n```\n{agent_name}\u002F\n  config.yaml\n  AGENTS.md\n  agents\u002F\n    {sub_agent_1_dir}\u002F\n      config.yaml\n    {sub_agent_2_dir}\u002F\n      config.yaml\n```\n\nDirectory names are free-form. A sub-agent's identity is the `name` in its\nown `config.yaml`, and that is what the parent lists in `tools.agents` —\n`{sub_agent_1_dir}` and `{sub_agent_1}` may differ.\n\n**Parent config.yaml:**\n```yaml\nspec_version: 1\nname: {agent_name}\ndescription: {description}\nexecutor:\n  type: omnigent\n  config:\n    harness: claude-sdk\ntools:\n  agents:\n    - {sub_agent_1}\n    - {sub_agent_2}\ninstructions: AGENTS.md\n```\n\n**Sub-agent config (agents\u002F{sub_agent_1_dir}\u002Fconfig.yaml):**\n```yaml\nspec_version: 1\nname: {sub_agent_1}\ndescription: {sub_agent_1_description}\nexecutor:            # any executor works here — only the parent needs omnigent\n  type: omnigent\n  config:\n    harness: claude-sdk\ninstructions: |\n  You are {sub_agent_1}. {sub_agent_1_instructions}\n```\n\n**Parent AGENTS.md should reference sub-agents:**\n```markdown\nYou have sub-agents you can delegate to:\n- **{sub_agent_1}** — {sub_agent_1_description}\n- **{sub_agent_2}** — {sub_agent_2_description}\n\nCall `sys_session_send(type=\"\u003Cname>\", input=\"\u003Ctask>\")` to dispatch a\ndeclared sub-agent. Emit multiple `sys_session_send` tool calls in the\nsame response to run them in parallel; results arrive via the inbox.\n```\n\n## Environment variable naming conventions\n\nWhen pinning credentials with `${ENV_VAR}`, map providers to their\nstandard env var names:\n- `openai` → `OPENAI_API_KEY`\n- `anthropic` → `ANTHROPIC_API_KEY`\n- `gemini` → `GEMINI_API_KEY` or `GOOGLE_API_KEY`\n- `groq` → `GROQ_API_KEY`\n- `deepseek` → `DEEPSEEK_API_KEY`\n- `xai` → `XAI_API_KEY`\n- `mistral` → `MISTRAL_API_KEY`\n- `databricks` → `DATABRICKS_TOKEN` (or an `auth.profile`)\n\n## Validation checklist\n\nBefore presenting the generated files to the user, verify (and if\n`validate_agent` is available, run it to confirm):\n- [ ] `spec_version: 1` is present\n- [ ] `name` is set and uses lowercase + hyphens\n- [ ] `executor.type` is one of `claude_sdk`, `agents_sdk`, `omnigent`\n- [ ] If `executor.type: omnigent`, `executor.config.harness` is set to a\n      valid harness\n- [ ] `instructions` (or `prompt`) points to a file that exists or is\n      inline text\n- [ ] When declaring `tools.agents`, the parent uses `executor.type:\n      omnigent`, and each entry is the declared `name` of a sub-agent under\n      `agents\u002F` (its directory name may differ; sub-agents may use any\n      executor)\n- [ ] `tools.builtins` names are from the known set (Step 2) — or, if\n      `list_builtin_tools` is available, were confirmed against it\n- [ ] Skill names use the `[a-z0-9-]+` pattern (they need not match their\n      directory names)\n",{"name":5,"description":7},"deploy\u002Froland-id-au-omniagent","83ee3dcd78468bbe3baf4bb264a0c99a53aa0fec","work\u002Fomnigent-main\u002Fomnigent\u002Fonboarding\u002Fagent\u002Fskills\u002Fbuild-omnigent","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent","https:\u002F\u002Fgithub.com\u002Froland-id-au","2026-09-16 03:15:34",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add roland-id-au\u002Fomniagent --skill build-omnigent",{"tool":36,"command":37},"bunx","bunx skills add roland-id-au\u002Fomniagent --skill build-omnigent",{"tool":39,"command":40},"pnpm","pnpm dlx skills add roland-id-au\u002Fomniagent --skill build-omnigent","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Fomnigent\u002Fonboarding\u002Fagent\u002Fskills\u002Fbuild-omnigent\u002FSKILL.md",true,[],[45,52,59,66,71,76],{"skillNo":46,"name":47,"nameZh":6,"description":48,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":49,"categories":50,"keywords":51,"publishedAt":19},"01M7CTDVNPM78AJ051GVDB44X5","deep-research","Research checklist and source-quality guidance for the archer e2e fixture.","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Ftests\u002Fresources\u002Fexamples\u002Farcher\u002Fskills\u002Fdeep-research\u002FSKILL.md",[],[],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":19},"01M7CTDS8SPN8P5GK1Q4SCSY7K","pi-e2e-xyz-greet-c4a8d5","Test greeter — pi skills e2e. The unique suffix c4a8d5 appears in the skill name so a string match in the agent's output proves Pi loaded this skill.","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Ftests\u002Fresources\u002Fagents\u002Fpi_skills_none\u002Fskills\u002Fpi-e2e-xyz-greet-c4a8d5\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M7CTDP4R9F06966W78WNZS2Y","pi-e2e-xyz-count-d2f6e1","Test counter — pi skills e2e. The unique suffix d2f6e1 appears in the skill name so a string match in the agent's output proves Pi loaded this skill.","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Ftests\u002Fresources\u002Fagents\u002Fpi_skills_none\u002Fskills\u002Fpi-e2e-xyz-count-d2f6e1\u002FSKILL.md",[],[],{"skillNo":67,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M7CTDKP4W1Q18ZCY0XANDSEG","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Ftests\u002Fresources\u002Fagents\u002Fpi_skills_list\u002Fskills\u002Fpi-e2e-xyz-greet-c4a8d5\u002FSKILL.md",[],[],{"skillNo":72,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7CTDFBMA3EBZNV7J380THGP","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Ftests\u002Fresources\u002Fagents\u002Fpi_skills_list\u002Fskills\u002Fpi-e2e-xyz-count-d2f6e1\u002FSKILL.md",[],[],{"skillNo":77,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7CTDCVTXTG72BKEV2HY7M6W","https:\u002F\u002Fgithub.com\u002Froland-id-au\u002Fomniagent\u002Fblob\u002Fdeploy\u002Froland-id-au-omniagent\u002Fwork\u002Fomnigent-main\u002Ftests\u002Fresources\u002Fagents\u002Fpi_skills_all\u002Fskills\u002Fpi-e2e-xyz-greet-c4a8d5\u002FSKILL.md",[],[]]