[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M77M37CTVPFP2Q5NAT412JS3":3,"skill-same:01M77M37CTVPFP2Q5NAT412JS3":44},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":13,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":6,"ownerDescription":28,"updatedAt":29,"install":30,"resources":43},"01M77M37CTVPFP2Q5NAT412JS3","investigate",null,"Delegate read-only investigation, debugging, audit, search, or code-understanding tasks to sub-agents; synthesize only from their structured reports.","hgxszhj\u002Fomnigent","hgxszhj","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F20251322?v=4","01M75Q4XCGWM2MCG58H8EZSB0F","01M77M1G2D2WP6EW4Y00K7W8HM",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\u002Fomnigent\u002Fresources\u002Fexamples\u002Fpolly\u002Fskills\u002Finvestigate\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: investigate\ndescription: Delegate read-only investigation, debugging, audit, search, or code-understanding tasks to sub-agents; synthesize only from their structured reports.\n---\n\n# investigate — delegated read-only work\n\nUse for any read-only task: investigation, debugging, audit, search, code\nunderstanding, architecture comparison, failure analysis, or answering a\nrepository-specific technical question.\n\n## Procedure\n1. Decompose the question into one or more bounded investigation tasks. Prefer\n   two independent lenses for ambiguous or high-stakes questions.\n2. Dispatch each task to `claude_code`, `codex`, or `pi`:\n   `sys_session_send(agent=\"claude_code\"|\"codex\"|\"pi\",\n   title=\"explore-\u003Ctask_slug>\", args={purpose: \"explore\", input: \"\u003Cquestion +\n   exact scope + evidence requested>\"})`. Use a task-based title such as\n   `explore-ci-flake`, never the raw vendor name. Use `purpose: \"search\"` only\n   when the task is primarily external\u002Fdocument search. Prefer `pi` when a\n   third lens or a non-Claude\u002FGPT model is wanted. Any worker takes an optional\n   `args.model` (`sys_list_models` shows what each worker can run; an invalid\n   model\u002Fworker combination fails loud at dispatch, and `model` only applies on\n   the dispatch that CREATES the session — a send that continues an existing\n   title rejects it).\n   Tell the worker to edit nothing and return file,\n   command, URL, or line evidence. Emit these `sys_session_send` calls in the\n   SAME turn — do not end a turn having only said you will dispatch.\n3. End your turn AFTER the dispatch tool calls are in flight (never before).\n   Do not inspect files, logs, terminals, docs, or connector output yourself\n   while the workers run.\n4. When workers finish, collect their completion results with\n   `sys_read_inbox`. Synthesize only from those inbox-delivered reports. Use\n   `sys_session_get_history` only to debug an empty or unclear worker result; if\n   reports conflict or are incomplete, dispatch a follow-up `explore` task\n   rather than resolving the conflict from your own direct inspection.\n5. If the investigation uncovers required code changes, switch to `fanout` \u002F\n   `cross-review`: dispatch an `implement` worker, then verify with the\n   opposite-vendor `review` worker.\n\n## Notes\n- The orchestrator may use its own tools only to create task packets, maintain\n  the registry, or check deterministic external status. It must not answer the\n  user's substantive question from its own direct file reads, shell output,\n  connector fetches, or terminal scrollback.\n- Keep task scopes narrow enough that each worker can return a concise report\n  with evidence. Broad investigations should be split into parallel subtasks.\n","# investigate — delegated read-only work\n\nUse for any read-only task: investigation, debugging, audit, search, code\nunderstanding, architecture comparison, failure analysis, or answering a\nrepository-specific technical question.\n\n## Procedure\n1. Decompose the question into one or more bounded investigation tasks. Prefer\n   two independent lenses for ambiguous or high-stakes questions.\n2. Dispatch each task to `claude_code`, `codex`, or `pi`:\n   `sys_session_send(agent=\"claude_code\"|\"codex\"|\"pi\",\n   title=\"explore-\u003Ctask_slug>\", args={purpose: \"explore\", input: \"\u003Cquestion +\n   exact scope + evidence requested>\"})`. Use a task-based title such as\n   `explore-ci-flake`, never the raw vendor name. Use `purpose: \"search\"` only\n   when the task is primarily external\u002Fdocument search. Prefer `pi` when a\n   third lens or a non-Claude\u002FGPT model is wanted. Any worker takes an optional\n   `args.model` (`sys_list_models` shows what each worker can run; an invalid\n   model\u002Fworker combination fails loud at dispatch, and `model` only applies on\n   the dispatch that CREATES the session — a send that continues an existing\n   title rejects it).\n   Tell the worker to edit nothing and return file,\n   command, URL, or line evidence. Emit these `sys_session_send` calls in the\n   SAME turn — do not end a turn having only said you will dispatch.\n3. End your turn AFTER the dispatch tool calls are in flight (never before).\n   Do not inspect files, logs, terminals, docs, or connector output yourself\n   while the workers run.\n4. When workers finish, collect their completion results with\n   `sys_read_inbox`. Synthesize only from those inbox-delivered reports. Use\n   `sys_session_get_history` only to debug an empty or unclear worker result; if\n   reports conflict or are incomplete, dispatch a follow-up `explore` task\n   rather than resolving the conflict from your own direct inspection.\n5. If the investigation uncovers required code changes, switch to `fanout` \u002F\n   `cross-review`: dispatch an `implement` worker, then verify with the\n   opposite-vendor `review` worker.\n\n## Notes\n- The orchestrator may use its own tools only to create task packets, maintain\n  the registry, or check deterministic external status. It must not answer the\n  user's substantive question from its own direct file reads, shell output,\n  connector fetches, or terminal scrollback.\n- Keep task scopes narrow enough that each worker can return a concise report\n  with evidence. Broad investigations should be split into parallel subtasks.\n",{"name":5,"description":7},"add-bubblewrap-installer-test","621465fa95293a450e35af83ea0cb89ec15b66e6","omnigent\u002Fresources\u002Fexamples\u002Fpolly\u002Fskills\u002Finvestigate","https:\u002F\u002Fgithub.com\u002Fhgxszhj\u002Fomnigent","https:\u002F\u002Fgithub.com\u002Fhgxszhj","我是一位网络技术工程测试人员","2026-06-24 06:54:25",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add hgxszhj\u002Fomnigent --skill investigate",{"tool":36,"command":37},"bunx","bunx skills add hgxszhj\u002Fomnigent --skill investigate",{"tool":39,"command":40},"pnpm","pnpm dlx skills add hgxszhj\u002Fomnigent --skill investigate","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\u002Fomnigent\u002Fresources\u002Fexamples\u002Fpolly\u002Fskills\u002Finvestigate\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":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":49,"categories":50,"keywords":51,"publishedAt":18},"01M77M4G5JEKDGGGZ3MS7TCDNG","deep-research","Research checklist and source-quality guidance for the archer e2e fixture.","https:\u002F\u002Fgithub.com\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\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":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":18},"01M77M4CXC4EJ0ZZSPM7XYBDBX","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\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\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":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":18},"01M77M49DRKYDZAFFNSZNKQ57G","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\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\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":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":18},"01M77M469MW3NK5FQ8G3PGPF7E","https:\u002F\u002Fgithub.com\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\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":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":18},"01M77M43H3TEFSAA9VG12BBQPR","https:\u002F\u002Fgithub.com\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\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":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":18},"01M77M404XAJMSA523HW29B3V3","https:\u002F\u002Fgithub.com\u002Fhgxszhj\u002Fomnigent\u002Fblob\u002Fadd-bubblewrap-installer-test\u002Ftests\u002Fresources\u002Fagents\u002Fpi_skills_all\u002Fskills\u002Fpi-e2e-xyz-greet-c4a8d5\u002FSKILL.md",[],[]]