[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MJF9HBW9DFTJFJGKBWM0BE":3,"skill-same:01M6MJF9HBW9DFTJFJGKBWM0BE":73},{"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":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":57,"updatedAt":58,"install":59,"resources":72},"01M6MJF9HBW9DFTJFJGKBWM0BE","ask",null,"Answer grounded questions about the codebase (onboarding \u002F Q&A) using the reviewer RAG + code graph. Use when the user asks how the code works, where something lives, or to explain a subsystem (\"where is auth\", \"how does X work\", \"explain the indexing flow\", \"как устроено…\", \"где у нас…\", \"объясни код\"). Requires a built base index + graph (reviewer MCP server). Not for reviewing PRs.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1559,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fmimfort\u002Frag_for_git\u002Fplugin\u002Fskills\u002Fask\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[5,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"reviewer","when","prs","for","q&a","answer","questions","indexing","use","work","MCP","explain","how","flow","server","graph","code","rag","auth","mcp server","2026-11-10 10:12:30",false,"---\nname: ask\ndescription: Answer grounded questions about the codebase (onboarding \u002F Q&A) using the reviewer RAG + code graph. Use when the user asks how the code works, where something lives, or to explain a subsystem (\"where is auth\", \"how does X work\", \"explain the indexing flow\", \"как устроено…\", \"где у нас…\", \"объясни код\"). Requires a built base index + graph (reviewer MCP server). Not for reviewing PRs.\n---\n\n# Ask (codebase Q&A)\n\nAnswer a free-text question about the codebase with a **grounded** response: every claim is\nbacked by a real `path:line` citation, never an invented path. Built on the reviewer MCP server\n(hybrid RAG + code graph). This skill reads and explains; it does NOT modify code or review PRs.\n\n**Always answer the user in Russian** (the project language), regardless of this file's language.\nTool calls, code identifiers, and `path:line` citations stay verbatim.\n\n## Inputs\n\n`$ARGUMENTS` — a free-text question about the code (e.g. \"where is authentication\",\n\"how does index freshness work\", \"explain the retrieval pipeline\").\n\n## Tools\n\n\u003C!-- include: _common\u002Ftool-usage.md -->\nUse the session-less tools above.\n\nPlus the harness file tools (`Read`, `Grep`, `Glob`) to read source from the local clone on disk.\n\n## Pipeline\n\n1. **Resolve repo\u002Fbranch.**\n   - `repo`: `git remote get-url origin` → strip a trailing `.git`, take the last two path\n     segments (`owner\u002Fname`). Pass `\"\"` to let the server use `DEFAULT_REPO` if origin is missing.\n   - `branch`:\n\n\u003C!-- include: _common\u002Fbranch-selection.md -->\n\n   **Freshness check (first code question of the session only).** After resolving repo\u002Fbranch and\n   ONLY on the first code question in this conversation — rely on conversation memory: if you have\n   already checked index freshness earlier in this session, skip this — run\n   `uvx --from rag-reviewer reviewer status \u003Cpath> --branch \u003Cbranch> --json` and read `drift`. If\n   `drift > 0`, emit exactly **one banner line**, in Russian:\n   «⚠ индекс отстаёт на N коммитов, ответ может не учитывать свежие изменения → `rag-reviewer:sync-codebase`».\n   Do NOT block, reindex, ask for confirmation, or call `sync_board` — this is warn-only. Cost ≈ 0\n   Voyage (reads `index_meta` + local git). **Fail-open:** any error → skip the banner silently\n   (Q&A is latency-sensitive).\n\n1.5. **Subsystem prior (cheap, optional).** Call `get_subsystem_summaries(repo, branch, query=\"\u003Cthe user's question>\")`.\n   At scale (summary count above the deploy threshold) this returns the top-k subsystems nearest the\n   question; on small repos it returns all (back-compat). If it returns summaries, use the one matching\n   the question's subsystem as a high-level orientation **before** `search_codebase` — this cuts\n   exploration steps for architectural \u002F \"how does subsystem X work\" questions. The summary is only a\n   prior: every `path:line` you cite in the answer still comes from real code (`search_codebase` \u002F\n   `Read`), never from the summary text.\n   **Fail-open:** empty \u002F unavailable → skip this step and proceed exactly as before.\n\n2. **Search.** Call `search_codebase(repo, \"\u003Cquestion>\", branch=…)`. Parse the\n   `path#fqn (path:start-end)` headers to get candidate symbols (`node_id`) and line ranges.\n   If the result is `(ничего не найдено)`, go to Fallback.\n\n   **Lazy expansion (no user prompt).** If the output ends with a cliff\u002Frails note reporting a\n   high-scoring tail beyond the cut AND the question looks broad, you MAY re-call `search_codebase`\n   once with a higher ceiling (pass `top_k=\u003Cbigger>`), then merge. Do this silently — never pause to\n   ask the user.\n\n3. **Expand (only as needed).** For an architectural \u002F \"how does X work\" question, DEFAULT to\n   skipping the graph tools (`related_symbols` \u002F `callers` \u002F `definition`) — the hybrid search\n   usually suffices; CLAUDE.md \u002F README are cheap priors to consult first. Only when the answer\n   genuinely needs call relationships, for the symbols most relevant to the question, call\n   `related_symbols` \u002F `callers` \u002F `definition` to follow the graph. Do NOT expand everything —\n   only what the answer requires. Stop once you can answer.\n\n4. **Confirm source.** `search_codebase` snippets are line-numbered, so when the returned snippet\n   already shows the exact code you cite, you may cite `path:line` directly from the tool output —\n   a separate `Read` is not required for grounding. Use `Read` only when the snippet was truncated\n   (`[...truncated]`) or you need surrounding context. Never cite a `path:line` not present in any\n   tool output.\n\n5. **Answer (adaptive), in Russian.**\n   - Default (focused question): a direct answer in 2–4 sentences, then an **Evidence** list —\n     each item `path:line` + a one-line \"why this is relevant\".\n   - Broad question (\"explain subsystem X\"): expand into sections — Краткий ответ \u002F Ключевые\n     символы \u002F Поток \u002F Связанные места — each claim still carrying a confirmed `path:line`.\n\n## Grounding contract (hard rule)\n\nCite ONLY paths that were returned by a tool AND confirmed by the tool's line-numbered output or a `Read`. Never invent or guess a\npath or line number. If you cannot ground a statement, say so explicitly instead of fabricating a\ncitation. This is the skill's acceptance criterion.\n\nA line-numbered `search_codebase` snippet that contains the cited code counts as grounding — an\nextra `Read` of the same lines is redundant.\n\n## Fallback (fail-open)\n\nIf the reviewer MCP server is unreachable or returns `(ничего не найдено)` \u002F `(граф недоступен)`\n(Postgres\u002FNeo4j\u002Findex down), degrade gracefully:\n- Use the harness `Grep`\u002F`Glob`\u002F`Read` over the local clone to locate and confirm code.\n- Tell the user (in Russian) that semantic\u002Fgraph search was unavailable and the answer comes from\n  a lexical search, so it may be less complete.\nNever abort — always return the best grounded answer you can.\n\n## Notes\n\n- Precondition: the base index + graph must be built (`reviewer index`). Graph precision depends on\n  the backend: SCIP → `IMPLEMENTS` + accurate `CALLS`; tree-sitter → `CALLS` by name only.\n- Read-only: this skill never edits code and never posts to GitHub. Posting a human-facing review\n  guide to a PR is a separate skill (PRI-119).\n","# Ask (codebase Q&A)\n\nAnswer a free-text question about the codebase with a **grounded** response: every claim is\nbacked by a real `path:line` citation, never an invented path. Built on the reviewer MCP server\n(hybrid RAG + code graph). This skill reads and explains; it does NOT modify code or review PRs.\n\n**Always answer the user in Russian** (the project language), regardless of this file's language.\nTool calls, code identifiers, and `path:line` citations stay verbatim.\n\n## Inputs\n\n`$ARGUMENTS` — a free-text question about the code (e.g. \"where is authentication\",\n\"how does index freshness work\", \"explain the retrieval pipeline\").\n\n## Tools\n\n\u003C!-- include: _common\u002Ftool-usage.md -->\nUse the session-less tools above.\n\nPlus the harness file tools (`Read`, `Grep`, `Glob`) to read source from the local clone on disk.\n\n## Pipeline\n\n1. **Resolve repo\u002Fbranch.**\n   - `repo`: `git remote get-url origin` → strip a trailing `.git`, take the last two path\n     segments (`owner\u002Fname`). Pass `\"\"` to let the server use `DEFAULT_REPO` if origin is missing.\n   - `branch`:\n\n\u003C!-- include: _common\u002Fbranch-selection.md -->\n\n   **Freshness check (first code question of the session only).** After resolving repo\u002Fbranch and\n   ONLY on the first code question in this conversation — rely on conversation memory: if you have\n   already checked index freshness earlier in this session, skip this — run\n   `uvx --from rag-reviewer reviewer status \u003Cpath> --branch \u003Cbranch> --json` and read `drift`. If\n   `drift > 0`, emit exactly **one banner line**, in Russian:\n   «⚠ индекс отстаёт на N коммитов, ответ может не учитывать свежие изменения → `rag-reviewer:sync-codebase`».\n   Do NOT block, reindex, ask for confirmation, or call `sync_board` — this is warn-only. Cost ≈ 0\n   Voyage (reads `index_meta` + local git). **Fail-open:** any error → skip the banner silently\n   (Q&A is latency-sensitive).\n\n1.5. **Subsystem prior (cheap, optional).** Call `get_subsystem_summaries(repo, branch, query=\"\u003Cthe user's question>\")`.\n   At scale (summary count above the deploy threshold) this returns the top-k subsystems nearest the\n   question; on small repos it returns all (back-compat). If it returns summaries, use the one matching\n   the question's subsystem as a high-level orientation **before** `search_codebase` — this cuts\n   exploration steps for architectural \u002F \"how does subsystem X work\" questions. The summary is only a\n   prior: every `path:line` you cite in the answer still comes from real code (`search_codebase` \u002F\n   `Read`), never from the summary text.\n   **Fail-open:** empty \u002F unavailable → skip this step and proceed exactly as before.\n\n2. **Search.** Call `search_codebase(repo, \"\u003Cquestion>\", branch=…)`. Parse the\n   `path#fqn (path:start-end)` headers to get candidate symbols (`node_id`) and line ranges.\n   If the result is `(ничего не найдено)`, go to Fallback.\n\n   **Lazy expansion (no user prompt).** If the output ends with a cliff\u002Frails note reporting a\n   high-scoring tail beyond the cut AND the question looks broad, you MAY re-call `search_codebase`\n   once with a higher ceiling (pass `top_k=\u003Cbigger>`), then merge. Do this silently — never pause to\n   ask the user.\n\n3. **Expand (only as needed).** For an architectural \u002F \"how does X work\" question, DEFAULT to\n   skipping the graph tools (`related_symbols` \u002F `callers` \u002F `definition`) — the hybrid search\n   usually suffices; CLAUDE.md \u002F README are cheap priors to consult first. Only when the answer\n   genuinely needs call relationships, for the symbols most relevant to the question, call\n   `related_symbols` \u002F `callers` \u002F `definition` to follow the graph. Do NOT expand everything —\n   only what the answer requires. Stop once you can answer.\n\n4. **Confirm source.** `search_codebase` snippets are line-numbered, so when the returned snippet\n   already shows the exact code you cite, you may cite `path:line` directly from the tool output —\n   a separate `Read` is not required for grounding. Use `Read` only when the snippet was truncated\n   (`[...truncated]`) or you need surrounding context. Never cite a `path:line` not present in any\n   tool output.\n\n5. **Answer (adaptive), in Russian.**\n   - Default (focused question): a direct answer in 2–4 sentences, then an **Evidence** list —\n     each item `path:line` + a one-line \"why this is relevant\".\n   - Broad question (\"explain subsystem X\"): expand into sections — Краткий ответ \u002F Ключевые\n     символы \u002F Поток \u002F Связанные места — each claim still carrying a confirmed `path:line`.\n\n## Grounding contract (hard rule)\n\nCite ONLY paths that were returned by a tool AND confirmed by the tool's line-numbered output or a `Read`. Never invent or guess a\npath or line number. If you cannot ground a statement, say so explicitly instead of fabricating a\ncitation. This is the skill's acceptance criterion.\n\nA line-numbered `search_codebase` snippet that contains the cited code counts as grounding — an\nextra `Read` of the same lines is redundant.\n\n## Fallback (fail-open)\n\nIf the reviewer MCP server is unreachable or returns `(ничего не найдено)` \u002F `(граф недоступен)`\n(Postgres\u002FNeo4j\u002Findex down), degrade gracefully:\n- Use the harness `Grep`\u002F`Glob`\u002F`Read` over the local clone to locate and confirm code.\n- Tell the user (in Russian) that semantic\u002Fgraph search was unavailable and the answer comes from\n  a lexical search, so it may be less complete.\nNever abort — always return the best grounded answer you can.\n\n## Notes\n\n- Precondition: the base index + graph must be built (`reviewer index`). Graph precision depends on\n  the backend: SCIP → `IMPLEMENTS` + accurate `CALLS`; tree-sitter → `CALLS` by name only.\n- Read-only: this skill never edits code and never posts to GitHub. Posting a human-facing review\n  guide to a PR is a separate skill (PRI-119).\n",{"name":5,"description":7},"main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002Fmimfort\u002Frag_for_git\u002Fplugin\u002Fskills\u002Fask",474,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-11 14:40:14",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill ask",{"tool":65,"command":66},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill ask",{"tool":68,"command":69},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill ask","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fmimfort\u002Frag_for_git\u002Fplugin\u002Fskills\u002Fask\u002FSKILL.md",true,[],[74,99,124,152,170,183],{"skillNo":75,"name":76,"nameZh":6,"description":77,"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":44},"01M716A6ETNB9B1SQ3GXZC75VR","deja-search","deja-vu memory — search the user's past AI coding sessions with the deja CLI. Holds what they already settled and never wrote into the repo (naming rules, banned or already-tried tools, required checks before merge, preferred values, who owns what). Use before your first edit in a task, before you call a change done or ready to merge, before you recommend a tool, library, design or reviewer, before re-debugging an error that may already be solved, when they say 'didn't we fix this before', 'what did we decide about X' or 'I already have that', and before telling them something on this machine does not exist.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-search\u002FSKILL.md",[],[81,82,83,84,85,86,87,88,89,90,91,25,92,93,24,94,95,96,97,98],"search","deja search","error","ready","never","memory","re","decide","fix","design","debugging","you","done","io","change","cli","merge","past",{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":44},"01M716A3J2W5SENSE9VGMXB375","deja-reflect","End-of-session reflection and memory curation with deja-vu. Review what was decided, what commands broke and what repaired them, and record durable project rules or decisions with deja remember so the next session starts with them.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-reflect\u002FSKILL.md",[],[106,107,108,109,86,110,111,112,113,114,115,116,117,94,118,119,120,121,122,123],"reflect","deja reflect","brew","remember","project rules","project","session","end","reflection","Rules","decisions","review","with","record","github","next","https","node",{"skillNo":125,"name":126,"nameZh":6,"description":127,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":128,"categories":129,"keywords":137,"publishedAt":44},"01M715T3RR4BAYSR57HPPTJ0PF","revid","Make, edit, export and post AI videos with Revid (revid.ai) through the Revid MCP server. Use whenever the user wants a short-form or long-form video made (from an idea, a script, a URL, an article, a PDF, a tweet, a product page, their own clips, or a song), wants to find what is going viral on TikTok, YouTube, Instagram or X before making something, wants to edit or re-export an existing Revid project, generate a standalone AI image or clip, make a series episode, or publish and schedule a finished video to connected social accounts. Covers setup (hosted MCP with OAuth, or an API key), the 36 tools, credits, polling and the rules for publishing. Routes to the more specific revid-* skills when one fits.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftmakersas\u002Frevid-skills\u002Fskills\u002Frevid\u002FSKILL.md",[130,133,136],{"categoryNo":131,"categoryName":132,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":134,"categoryName":135,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[126,138,139,140,141,142,111,143,87,144,145,34,146,25,147,148,43,149,150,151],"Video","short-form","standalone","PDF","viral","tiktok","short","social","song","URL","image","schedule","api","publish",{"skillNo":153,"name":154,"nameZh":6,"description":155,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":156,"categories":157,"keywords":158,"publishedAt":44},"01M71569ZM2X3H5RBFK8PVKMK2","review-study","Review a QuestionPunk survey before launch for wording, answer choices, branching, and missing settings. Use for check my survey or review this draft before publishing.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Freview-study\u002FSKILL.md",[],[159,160,117,27,29,161,162,163,32,164,165,166,167,168,169],"study","review study","missing","survey","publishing","check","draft","settings","launch","my","branching",{"skillNo":171,"name":172,"nameZh":6,"description":173,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":174,"categories":175,"keywords":176,"publishedAt":44},"01M71567GV0C35YQBY0DPGR86B","manage-study","Find QuestionPunk studies, check status, publish an explicitly requested draft, or retrieve its participant link. Use for find my survey, publish this study, or how many people responded.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Fmanage-study\u002FSKILL.md",[],[177,159,178,27,162,32,179,164,165,151,36,180,181,182,168],"manage study","manage","people","find","link","status",{"skillNo":184,"name":185,"nameZh":6,"description":186,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":187,"categories":188,"keywords":189,"publishedAt":44},"01M71565A8TCA01AZQZRGRYMFZ","create-study","Create a saved QuestionPunk survey draft from a research brief or edit specified questions. Use for requests such as make a customer feedback survey or shorten my onboarding study.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Fcreate-study\u002FSKILL.md",[],[190,159,191,192,193,194,27,195,162,196,32,197,198,165,199,200,201,30,202,168],"create study","create","brief","as","feedback","customer feedback","research brief","shorten","saved","research","requests","customer","edit"]