ruc-nlpir
- 91 skills
- 0 followers
- 23 repo stars
- yesterday last updated
- ▌ Appworld Spotify Set Based Aggregate Verification · ruc-nlpirAppWorld Spotify cleanup or bulk tasks where an aggregate state such as "an album is downloaded iff every song in it is downloaded" determines keep or remove decisions. Use when printed ID lists could be misread and per-item downloaded counts must be exact.
- ▌ Appworld Filesystem Bill Aggregation Mixed Formats · ruc-nlpirAppWorld file_system tasks that aggregate bills or financial documents over a date range. Use when a directory mixes .txt and .pdf bill files and an in-scope period appears only as a PDF, or when file_system.show_file returns a binary placeholder and the plan would skip PDF files as duplicates.
- ▌ Verify Embedded Version After Fetching Source · ruc-nlpirUse when building a version-pinned compiler or tool from a Git tag or source archive, and the verifier or user may check the version the built tool reports.
- ▌ Dataset Domain Token Counts Via Metadata Config · ruc-nlpirUse when a HuggingFace dataset task asks for token counts or filtering by a named domain and the dataset has multiple configs, only one of which contains domain and token-text columns.
- ▌ Short Answer Exact Extraction · ruc-nlpirUse for exact-match short-answer QA; return only the minimal slot value with the expected surface form, anchored to the strongest context support.
- ▌ Answer Source Form Verification · ruc-nlpirDetermine the authoritative source, answer type, and exact surface form before committing an exact-match short answer. Use when the retrieved context interleaves a speaker's guess with an 'A:' answer, when the prompt labelled 'Question' is a bare declarative statement with an unresolved pronoun, or when the prompt's 'one of these [plural]' phrasing dictates a singular noun with article.
- ▌ Appworld Paginated API Page Limit · ruc-nlpirAppWorld paginated list/search endpoints that validate page_limit. Use when a bulk fetch plan may set page_limit above the API maximum, or when a call returns HTTP 422 about a page_limit constraint.
- ▌ Appworld Simple Note Search Query · ruc-nlpirAppWorld simple_note retrieval tasks where a target note must be found among many routine or pinned notes. Use when broad search_notes output is dominated by unrelated recurring notes.
- ▌ Appworld Spotify Liked Vs Library · ruc-nlpirAppWorld Spotify tasks that ask about the album library, liked playlists, "my playlists", or "my song library and playlists". Use when wording distinguishes album library from liked albums, uses a relative clause such as "playlists which I have liked", or when a destructive cleanup scope may require liked playlists in addition to owned library playlists.
- ▌ Appworld Spotify Show Song Fields · ruc-nlpirAppWorld Spotify tasks that filter library or playlist songs by release date. Use when show_song_library or show_playlist_library returns song IDs without release dates and the plan is to call show_song per song.
- ▌ Appworld Supervisor Complete Task · ruc-nlpirAppWorld tasks that require an explicit supervisor.complete_task call after the requested API action has succeeded. Use when the terminal action returned success and a read-back already confirms the target state while a valid access token is still held.
- ▌ Final Answer Commit Verification · ruc-nlpirVerify representation, evidence, and referent before committing a graded answer. Use when a task is graded by exact string match and a semantically correct answer still fails because its form is over-elaborated, its verbatim evidence was ignored, or it names the wrong referent among shared candidates.
- ▌ Appworld API Preflight Verification · ruc-nlpirAppWorld tasks requiring bulk or mutating API work across records, playlists, notes, or pages. Use when the solver is about to loop over API results; verify the exact task target set, endpoint pagination limits, and actual response schema before bulk operations.
- ▌ Claude Code Unrecognized Model · ruc-nlpirDetect a fatal Claude Code harness model-configuration fault and abort early. Trigger when a run's stderr contains `claude-code:unrecognized_model` or the configured `model` field names an ID the CLI does not recognize (e.g. `deepseek-flash`), so the run would otherwise stall to its full timeout with no task artifacts.
- ▌ Venmo Relationship Target Derivation · ruc-nlpirUse for AppWorld/Venmo tasks that mutate records qualified by a social relationship plus a time window (e.g. "payments I received from my friends"), where the target set must be derived by cross-referencing endpoints before bulk comment/like mutations.
- ▌ SQL Correlated Subquery Rewrite · ruc-nlpirOptimize a SQL query over a large unindexed table by collapsing repeated correlated subqueries or per-group "top-N" lookups into a single computation, then verify output equivalence. Trigger when the task needs one rewritten SQL deliverable and schema inspection shows no indexes.
- ▌ Appworld Authenticated API Access Token · ruc-nlpirAppWorld tasks whose app APIs require authentication. Use when an app API call returns HTTP 401 or its docs list a required access_token parameter, and after an app login call has returned an access_token.
- ▌ Appworld Execute Recovery And Verification · ruc-nlpirAppWorld tasks when an execute block fails partway or mutates API state. Use when retrying a failed script, following/liking records, persisting tokens, or logging into phone/venmo; preserve original task filters, re-fetch mutated values before answering, avoid disallowed built-ins, and verify app-specific login identifiers.
- ▌ Install Required Runtime And Artifact · ruc-nlpirUse when a terminal coding task explicitly names a required language/runtime and a required implementation or output file path, and the verifier tests assert that file exists.
- ▌ Appworld Spotify Queue Played So Far Cyclic · ruc-nlpirAppWorld Spotify music-player queue tasks that ask which songs have been "played so far". Use when show_song_queue returns a cyclic queue with the current song not at the first or last position, or when a position-based subset of the queue is about to be used as the played set.
- ▌ Fpbase Spectral Peaks Read From States · ruc-nlpirUse when a task selects or validates donor/acceptor fluorescent proteins by peak excitation or emission through the FPbase API and direct top-level ex_max/em_max reads return None.
- ▌ Repair Verifier Bootstrap Before Rerun · ruc-nlpirUse when a task solution produces the expected output files and local checks pass, but the verifier still reports zero reward and the captured test output shows the test harness failing before task-specific assertions.
- ▌ Appworld Spotify Playlist Duration Not Count · ruc-nlpirAppWorld Spotify tasks that select or start a playlist whose total duration must cover a fixed activity length. Use when playlist search or library endpoints return song_ids without durations and the plan might treat len(song_ids) as a proxy for total time.
- ▌ Verify Build Prerequisites Before Compiling · ruc-nlpirUse when a task asks for a non-native compiled artifact, such as a cross-compiled MIPS ELF, in a minimal or unfamiliar container that already contains a source tree and Makefile.
- ▌
- ▌
- ▌ Appworld · ruc-nlpirUse when solving AppWorld tasks with the appworld_execute MCP tool. Covers safe API discovery, action execution, and supervisor completion.
- ▌ Hermes Agent · ruc-nlpirUse the FlashRSI Hermes Snapshot's headless Python runtime, tools, sessions, and compression paths.
- ▌
- ▌ Answer Verifier · ruc-nlpirUse after drafting a short answer to a supplied-context question. Corrects the draft to the exact span required by the question focus and returns only that answer.
- ▌ Terminal 3 · ruc-nlpirUse for terminal task execution in this experiment. General playbook for filesystem, data, build, database, and simulation tasks.
- ▌
- ▌
- ▌ Exact Match Answer Form · ruc-nlpirUse for exact-match short-answer QA to copy the shortest required source span; choose name form by question phrasing, avoid repeating question nouns, prefer singular source category nouns for plural-sounding multi-entity questions, and preserve source numeral wording.
- ▌ Answer Form Verification · ruc-nlpirUse before finalizing a short answer when scoring may be exact or span-based. Verifies that the candidate copies the full source span the question requires, including formal names, minimal entity names, category head nouns, complete predicate frames, extent qualifiers, and direct anaphor resolution.
- ▌ Scientific Impl · ruc-nlpirUse for tasks that implement a numerical, statistical, or sampling algorithm with formal tests against known truth — "implement a sampler / estimator / integrator", "compare results to known distributions", "validate inputs and reject invalid cases". Covers R/Python scientific code with modular functions and PASS/FAIL test output.
- ▌ SQL Optimization · ruc-nlpirUse when the task is to make a SQL query faster while producing identical output — "optimize this query", "make it efficient", "same output produced". Covers schema inspection, rewrite patterns, and byte-identical verification for SQLite and other SQL dialects.
- ▌ Short Answer Exact Match · ruc-nlpirApply when a short-answer or multi-hop QA prompt instructs "return only the short answer" (or similar) and scoring is exact match. Emit the minimal answer label — stripped of parentheticals, leading qualifiers, and trailing category words — but preserve a person's full canonical name as given in the source; when the answer already appears verbatim in the supplied context, resolve from that context without tool searching or over-deliberation, and make sure it appears as visible text.
- ▌ Short Answer Exact Match 2 · ruc-nlpirReturn the minimal exact-match answer for short-answer QA. Use when a prompt instructs "return only the short answer" (or otherwise implies exact-string grading) and asks for a name, nickname, alias, title, symbol/code, or person (a person named by a first name or full name — answer with the bare surname); when the context offers a full title and a shorter common name, variant spellings/aliases of the target, or overlapping group terms whose membership must cover every listed item; when a title/role is shared across people and a "current"/"former" qualifier disambiguates; or when exact match hinges on verbatim spacing/punctuation, number inflection, or a nickname-vs-formal-name choice.
- ▌ Build From Source · ruc-nlpirUse for tasks that require compiling, building, cross-compiling, porting, or repairing an existing codebase — "build X from source", "fix the build", "port to Y", "get the tests green". A failure-first protocol that avoids thrashing on long builds.
- ▌ Short Answer Slot Anchoring · ruc-nlpirExact-match short-answer QA; return the exact minimal slot surface form required by prompt and context.
- ▌ Appworld API Doc Retrieval · ruc-nlpirUse for AppWorld/FlashRSI tasks where the session already lists flashrsi:* helper skills or where you need one or a few API signatures from an app with a large API surface (e.g. Spotify, Venmo); use a listed helper skill and request docs for the specific API name instead of dumping the whole app.
- ▌ Appworld File System Access · ruc-nlpirUse for AppWorld tasks that reference files or directories (e.g. "~/bills/") whose data lives inside the simulated file_system app; skip host-shell filesystem probing and read via the file_system app API after fetching credentials.
- ▌ Minimal Answer Verification · ruc-nlpirCalibrate and verify minimal outputs for automatically-graded short answers. Use when the answer span appears verbatim in the supplied context, when a context offers both specific and generic phrasings, or when a yes/no claim shares a salient keyword with evidence but the actual outcome must be matched.
- ▌ Hotpotqa · ruc-nlpirUse for every supplied-context question-answering task in this HotpotQA experiment.
- ▌ Searchqa · ruc-nlpirUse for every supplied-context question-answering task in this SearchQA experiment.
- ▌ Searchqa 2 · ruc-nlpirUse for every supplied-context question-answering task where the answer is in, or directly derivable from, the supplied snippets or paragraphs.
- ▌ Searchqa 3 · ruc-nlpirSlot-first, type-routed, evidence-grounded question answering for SearchQA tasks. Use for every question where the supplied context contains search snippets, web pages, or retrieved documents.
- ▌ Git 2 · ruc-nlpirGit commits, branches, rebases, merges, conflict resolution, history recovery, team workflows, and the commands needed for safe day-to-day version control. Use when (1) the task touches Git, a repository, commits, branches, merges, rebases, or pull requests; (2) history safety, collaboration, or recovery matter; (3) the agent should automatically apply Git discipline instead of improvising.
- ▌ Docker 2 · ruc-nlpirDocker containers, images, Compose stacks, networking, volumes, debugging, production hardening, and the commands that keep real environments stable. Use when (1) the task touches Docker, Dockerfiles, images, containers, or Compose; (2) build reliability, runtime behavior, logs, ports, volumes, or security matter; (3) the agent needs Docker guidance and should apply it by default.
- ▌ Terminal 2 · ruc-nlpirLocal shell copilot for command planning, safe execution, preview-first workflows, output summarization, privacy-aware history controls, and step-by-step terminal help. Use whenever the user wants to run terminal commands, inspect files, debug shell issues, automate local tasks, or translate natural language into shell actions. Prefer safe preview before mutation. Require explicit confirmation for destructive commands. Local-only.
- ▌ Productivity 2 · ruc-nlpirPlan, focus, and complete work with energy management, time blocking, goals, projects, tasks, habits, reviews, priorities, and context-specific productivity systems; use when (1) the user needs help with productivity, focus, time management, planning, priorities, goals, projects, tasks, habits, or reviews; (2) they want a reusable structure or workspace for organizing work; (3) ongoing work should be routed through a dedicated productivity framework.
- ▌
- ▌ File Manager 3 · ruc-nlpirOpenClaw automated file management assistant for batch file operations, smart classification, duplicate file cleanup, file renaming, directory sync, and file workflow automation. Use when the user needs to organize files, batch rename, clean duplicates, sync directories, or automate file workflows.
- ▌ Web Search Plus 3 · ruc-nlpirUnified search skill with Intelligent Auto-Routing. Uses multi-signal analysis to automatically select between Serper (Google), Tavily (Research), Querit (Multilingual AI Search), Exa (Neural), Perplexity (AI Answers), You.com (RAG/Real-time), and SearXNG (Privacy/Self-hosted) with confidence scoring.
- ▌ Web Search Plus 4 · ruc-nlpir bundleUnified search skill with Intelligent Auto-Routing. Uses multi-signal analysis to automatically select between Serper (Google), Tavily (Research), Querit (Multilingual AI Search), Exa (Neural), Perplexity (AI Answers), You.com (RAG/Real-time), and SearXNG (Privacy/Self-hosted) with confidence scoring.
- ▌ Obsidian 2 · ruc-nlpirWork with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
- ▌ Gitlab CLI Skills 2 · ruc-nlpirComprehensive GitLab CLI (glab) command reference and workflows for all GitLab operations. Use when working with merge requests, CI/CD pipelines, issues, releases, repositories, authentication, variables, labels, milestones, snippets, or any glab command. Covers 37+ sub-commands including glab mr, glab ci, glab issue, glab repo, glab release, glab variable, and more.
- ▌ Word DOCX 2 · ruc-nlpirCreate, inspect, and edit Microsoft Word documents and DOCX files with reliable styles, numbering, tracked changes, tables, sections, and compatibility checks.
- ▌ Planning With Files 2 · ruc-nlpirImplements Manus-style file-based planning to organize and track progress on complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when asked to plan out, break down, or organize a multi-step project, research task, or any work requiring >5 tool calls. Supports automatic session recovery after /clear.
- ▌ Excel XLSX 2 · ruc-nlpirCreate, inspect, and edit Microsoft Excel workbooks and XLSX files with reliable formulas, dates, types, formatting, recalculation, and template preservation.
- ▌ Paddleocr Doc Parsing 2 · ruc-nlpirUse this skill to extract structured Markdown/JSON from PDFs and document images—tables with cell-level precision, formulas as LaTeX, figures, seals, charts, headers/footers, multi-column layout and correct reading order. Trigger terms: 文档解析, 版面分析, 版面还原, 表格提取, 公式识别, 多栏排版, 扫描件结构化, 发票, 财报, 复杂 PDF, PDF转Markdown, 图表, 阅读顺序; reading order, formula, LaTeX, layout parsing, structure extraction, PP-StructureV3, PaddleOCR-VL.
- ▌ File Manager 4 · ruc-nlpirOpenClaw自动化文件管理助手,用于批量文件操作、智能分类、重复文件清理、文件重命名、目录同步等任务。当用户需要整理文件、批量重命名、清理重复文件、同步目录或自动化文件工作流时使用此技能。
- ▌ Imap Smtp Email 2 · ruc-nlpirRead and send email via IMAP/SMTP. Check for new/unread messages, fetch content, search mailboxes, mark as read/unread, and send emails with attachments. Supports multiple accounts.
- ▌ File Manager 2 · ruc-nlpirOpenClaw automated file management assistant for batch file operations, smart classification, duplicate file cleanup, file renaming, directory sync, and file workflow automation. Use when the user needs to organize files, batch rename, clean duplicates, sync directories, or automate file workflows.
- ▌ Web Search Plus 2 · ruc-nlpirUnified search skill with Intelligent Auto-Routing. Uses multi-signal analysis to automatically select between Serper (Google), Tavily (Research), Querit (Multilingual AI Search), Exa (Neural), Perplexity (AI Answers), You.com (RAG/Real-time), and SearXNG (Privacy/Self-hosted) with confidence scoring.
- ▌ Planning With Files · ruc-nlpir bundleImplements Manus-style file-based planning to organize and track progress on complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when asked to plan out, break down, or organize a multi-step project, research task, or any work requiring >5 tool calls. Supports automatic session recovery after /clear.
- ▌ Arbor Agent Tools · ruc-nlpir bundleDeterministic helper layer for emulating Arbor tools in Codex or Claude Code when native TreeView, TreeAddNode, TreeSetMeta, RunExecutor, GitMergeBranch, or report tooling is unavailable. Use for local state management, eval score capture, executor prompt generation, merge checks, and skill forward tests.
- ▌ Arbor Agent Ideate · ruc-nlpir bundleStrict IDEATE-stage skill for Arbor. Use immediately after TreeView(format="constraints") when drafting Idea Tree nodes, enforcing the idea_drafting and first_principles_probe behavior, depth-aware idea levels, four-line TreeAddNode hypotheses, conflict checks, and self-filtering against shallow tweaks.
- ▌ Arbor Agent Search · ruc-nlpir bundleRelated-work and novelty annotation phase for Arbor. Use after a node has been experimentally validated, especially before merge decisions, to emulate SearchIdeaContext/SearchIdeaContextParallel/SearchStatus, validated-node gating, background SearchAgent behavior, and structured prior-art summaries.
- ▌ Arbor Agent Executor · ruc-nlpir bundleExecutor-dispatch phase for Arbor. Use when implementing an Idea Tree node through RunExecutor or RunExecutorParallel semantics: isolated git worktree, executor prompt construction, eval metadata injection, RunTraining policy, smoke/full evaluation, report parsing, artifact persistence, tree update, and insight propagation.
- ▌ Arbor Research Agent · ruc-nlpir bundlePublic entrypoint for the Arbor skill suite. Use when a user wants to run an Arbor-style autonomous research or optimization workflow from a natural-language goal, including initial clarification of objective, target project, data, metric, evaluation, permissions, budget, run mode, and then automatic bootstrapping into arbor-agent-orchestrator and phase skills.
- ▌ Arbor Agent Merge Eval · ruc-nlpir bundleMerge and evaluation discipline for Arbor. Use for TreeSetMeta metadata, B_dev/B_test separation, eval command templates, score parsing, GitMergeBranch behavior, protected paths, required outputs, metric_direction, trunk/test score updates, medal detection, and final evaluation before stopping.
- ▌ Arbor Agent Coordinator · ruc-nlpir bundleCoordinator phase for Arbor: persistent ReAct loop, Idea Tree state, INIT/OBSERVE/IDEATE/SELECT/DISPATCH/DECIDE protocol, tool mapping, cycle caps, and coordinator-only behavior. Use after setup/intake and before phase-specific ideation, executor, merge, search, or report skills.
- ▌ Arbor Agent Orchestrator · ruc-nlpir bundleTop-level controller for recreating the open-source AutoResearch workflow as a suite of skills. Use when the user asks to run, emulate, extract, validate, or refine Arbor/AutoResearch behavior, especially when a coordinator must load phase skills for setup, ideation, executors, merge evaluation, novelty search, plugins, resume, and reports.
- ▌ Arbor Agent Setup Intake · ruc-nlpir bundleSetup, intake, preflight, and launch-contract phase for open-source Arbor runs. Use when confirming a target project, metric, baseline, dev/test split, config/plugin settings, branch guard, session directory, or when translating a user goal into the precise contract consumed by the coordinator.
- ▌ Arbor Agent Resume Report · ruc-nlpir bundleResume, checkpoint, event, dashboard, finalization, and report phase for Arbor. Use when continuing interrupted sessions, handling running-node requeue, writing checkpoints/messages, consuming events.jsonl/run_stats, recovering best submissions, generating REPORT.md, or summarizing durable artifacts.
- ▌ Arbor Agent Plugins Hitl Budget · ruc-nlpir bundleDomain adaptation, human-in-the-loop, and budget policy phase for Arbor. Use when a run mentions plugins, plugin profiles, mle_kaggle, eval_contract, protected_paths, required_outputs, lifecycle hooks, convergence, budget_policy, RunTraining stages, AskUser, or interaction modes auto/direction/review/collaborative.
- ▌ Git · ruc-nlpirGit commits, branches, rebases, merges, conflict resolution, history recovery, team workflows, and the commands needed for safe day-to-day version control. Use when (1) the task touches Git, a repository, commits, branches, merges, rebases, or pull requests; (2) history safety, collaboration, or recovery matter; (3) the agent should automatically apply Git discipline instead of improvising.
- ▌
- ▌ Docker · ruc-nlpirDocker containers, images, Compose stacks, networking, volumes, debugging, production hardening, and the commands that keep real environments stable. Use when (1) the task touches Docker, Dockerfiles, images, containers, or Compose; (2) build reliability, runtime behavior, logs, ports, volumes, or security matter; (3) the agent needs Docker guidance and should apply it by default.
- ▌ Terminal · ruc-nlpirLocal shell copilot for command planning, safe execution, preview-first workflows, output summarization, privacy-aware history controls, and step-by-step terminal help. Use whenever the user wants to run terminal commands, inspect files, debug shell issues, automate local tasks, or translate natural language into shell actions. Prefer safe preview before mutation. Require explicit confirmation for destructive commands. Local-only.
- ▌ Obsidian · ruc-nlpir bundleWork with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
- ▌ Word DOCX · ruc-nlpir bundleCreate, inspect, and edit Microsoft Word documents and DOCX files with reliable styles, numbering, tracked changes, tables, sections, and compatibility checks.
- ▌ Excel XLSX · ruc-nlpir bundleCreate, inspect, and edit Microsoft Excel workbooks and XLSX files with reliable formulas, dates, types, formatting, recalculation, and template preservation.
- ▌ Productivity · ruc-nlpirPlan, focus, and complete work with energy management, time blocking, goals, projects, tasks, habits, reviews, priorities, and context-specific productivity systems; use when (1) the user needs help with productivity, focus, time management, planning, priorities, goals, projects, tasks, habits, or reviews; (2) they want a reusable structure or workspace for organizing work; (3) ongoing work should be routed through a dedicated productivity framework.
- ▌ File Manager · ruc-nlpir bundleOpenClaw自动化文件管理助手,用于批量文件操作、智能分类、重复文件清理、文件重命名、目录同步等任务。当用户需要整理文件、批量重命名、清理重复文件、同步目录或自动化文件工作流时使用此技能。
- ▌ Web Search Plus · ruc-nlpirUnified search skill with Intelligent Auto-Routing. Uses multi-signal analysis to automatically select between Serper (Google), Tavily (Research), Querit (Multilingual AI Search), Exa (Neural), Perplexity (AI Answers), You.com (RAG/Real-time), and SearXNG (Privacy/Self-hosted) with confidence scoring.
- ▌ Imap Smtp Email · ruc-nlpir bundleRead and send email via IMAP/SMTP. Check for new/unread messages, fetch content, search mailboxes, mark as read/unread, and send emails with attachments. Supports multiple accounts.
- ▌ Gitlab CLI Skills · ruc-nlpirComprehensive GitLab CLI (glab) command reference and workflows for all GitLab operations. Use when working with merge requests, CI/CD pipelines, issues, releases, repositories, authentication, variables, labels, milestones, snippets, or any glab command. Covers 37+ sub-commands including glab mr, glab ci, glab issue, glab repo, glab release, glab variable, and more.
- ▌ Paddleocr Doc Parsing · ruc-nlpirUse this skill to extract structured Markdown/JSON from PDFs and document images—tables with cell-level precision, formulas as LaTeX, figures, seals, charts, headers/footers, multi-column layout and correct reading order. Trigger terms: 文档解析, 版面分析, 版面还原, 表格提取, 公式识别, 多栏排版, 扫描件结构化, 发票, 财报, 复杂 PDF, PDF转Markdown, 图表, 阅读顺序; reading order, formula, LaTeX, layout parsing, structure extraction, PP-StructureV3, PaddleOCR-VL.