550 个 · 按相关度排序
Sweep recent sessions across every agent CLI recorded in Whalesong and surface the ones worth a look — errors, runaway loops, token blowouts, stalled gaps. Use for "what have my agents been doing", morning reviews, or finding the session that went wrong.
中文简介正在准备中。当前显示原文说明。
Multi-agent autonomous startup system for Claude Code. Triggers on "Loki Mode". Orchestrates 100+ specialized agents across engineering, QA, DevOps, security, data/ML, business operations, marketing, HR, and customer success. Takes PRD to fully deployed, revenue-generating product with zero human intervention. Features Task tool for subagent dispatch, parallel code review with 3 specialized reviewers, severity-based issue triage, distributed task queue with dead letter handling, automatic deployment to cloud providers, A/B testing, customer feedback loops, incident response, circuit breakers, and self-healing. Handles rate limits via distributed state checkpoints and auto-resume with exponential backoff. Requires --dangerously-skip-permissions flag.
中文简介正在准备中。当前显示原文说明。
Autonomous agents are AI systems that can independently decompose goals, plan actions, execute tools, and self-correct without constant human guidance. The challenge isn't making them capable - it's making them reliable. Every extra decision multiplies failure probability. This skill covers agent loops (ReAct, Plan-Execute), goal decomposition, reflection patterns, and production reliability. Key insight: compounding error rates kill autonomous agents. A 95% success rate per step drops to 60% b
中文简介正在准备中。当前显示原文说明。
Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.
中文简介正在准备中。当前显示原文说明。
Create PostHog error tracking suppression rules to drop high-volume, low-value errors at ingestion. Use when the user asks "stop capturing this error", "drop browser extension errors", "ignore ResizeObserver loops", "suppress bot-driven errors", or wants to reduce ingestion cost from noisy unactionable errors. Identifies suppression candidates, scopes the filter tightly, decides between full suppression and sampling, and confirms the rule before creating it. Suppressed errors are dropped permanently — this skill defaults to caution.
中文简介正在准备中。当前显示原文说明。
Focused Signals scout for PostHog projects using error tracking. Watches `$exception` bursts, stuck loops, multi-fingerprint clusters, status regressions, and stack-trace activity-name patterns. Emits findings only when they clear the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-* fleet — no dependencies on other skills.
中文简介正在准备中。当前显示原文说明。
Iterate on a PR until actionable CI passes and high/medium review feedback is addressed. Use for PR CI failures, review feedback, or green-check loops; do not wait for human approval, draft status, or merge gates.
中文简介正在准备中。当前显示原文说明。
High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.
中文简介正在准备中。当前显示原文说明。
End-to-end Hi-C analysis workflow from FASTQ to compartments, TADs, and loops, with the decision of WHICH features the sequencing depth can support. Covers pairtools read-pair processing and library QC, cooler matrices, ICE balancing and distance-decay expected, A/B compartments, TAD boundaries, loop calling, and the routing of HiChIP/PLAC-seq/Capture Hi-C to protein-directed loop callers. Use when processing Hi-C data end to end, deciding a resolution for a given depth, or choosing between bulk-Hi-C and protein-directed loop calling.
中文简介正在准备中。当前显示原文说明。
Design AI feedback loops for continuous improvement. TRIGGERS - Use when user needs help with ai-feedback-loop related tasks.
中文简介正在准备中。当前显示原文说明。
Balances Hi-C contact matrices (ICE via cooler.balance_cooler, KR/SCALE/VC context), computes distance-decay expected with cooltools (expected_cis per-diagonal P(s), expected_trans scalar), builds observed/expected (O/E) matrices, and diagnoses polymer state from the P(s) log-derivative. Covers the within-matrix-vs-cross-sample distinction (balancing is NOT a normalizer), the equal-visibility assumption that CNV/aneuploidy violates (use raw counts for copy-number), cis-only balancing, mad_max/blacklist masking before balancing, multiplicative cooler weights vs divisive juicer weights, and the resolution-vs-depth budget. Use when balancing a .cool/.mcool, computing expected or P(s), making O/E matrices for compartments/loops, deciding ICE vs KR vs SCALE, choosing a resolution for a given depth, or troubleshooting NaN/all-NaN balanced matrices; route cross-sample comparison to hic-differential.
中文简介正在准备中。当前显示原文说明。
Bulk-query Ensembl BioMart (and other BioMart instances) for cross-database ID mapping, gene/transcript/exon coordinates, and ortholog tables. Use when batch-converting Ensembl IDs to other namespaces (HGNC, RefSeq, UniProt, Entrez), pulling gene coordinate tables for thousands of genes, building ortholog wide-tables across species, or replacing slow Ensembl REST loops with one-shot bulk export. Encodes BioMart's XML query format, R biomaRt vs Python pybiomart trade-off, mart-vs-dataset hierarchy, and the URL endpoint that's BioMart-specific (separate from rest.ensembl.org).
中文简介正在准备中。当前显示原文说明。
Iteratively detects and auto-fixes technical debt in a codebase using three quality criteria from debt-analysis references: conditional branch simplification, encapsulation, and separation of concerns. Applies fixes directly to code, runs existing tests to verify no regressions, re-analyzes, and loops until all high-severity issues are resolved or max iterations (5) reached. Use this skill when you want to automatically improve code quality with real code changes, reduce technical debt systematically, or clean up code to meet quality standards. Trigger on phrases like "fix technical debt", "clean up code quality", "refactor and fix issues", "improve code quality automatically", "apply debt fixes", "fix code problems", "eliminate debt".
中文简介正在准备中。当前显示原文说明。
High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.
中文简介正在准备中。当前显示原文说明。
Read-only consolidated snapshot of an SDD Composy workspace: lifecycle stage, task contracts, trace integrity, loops, fleets, and the next action, without changing any file. Use for 'em que pé está o SDD', 'status do fluxo', 'qual a próxima ação', 'where are we in the workflow'. Do NOT use for git, CI, infrastructure, or other plugins' status, or to change state.
中文简介正在准备中。当前显示原文说明。
Identifies and fixes common Rust anti-patterns. Use during code review, when seeing excessive clones/unwraps/index loops, or when code fights the borrow checker instead of working with it.
中文简介正在准备中。当前显示原文说明。
Automated PR review and CI auto-fix for GitHub and GitLab using the Composio CLI. Pulls diffs, fetches failing job logs, posts review comments, and loops fix commits until checks go green.
中文简介正在准备中。当前显示原文说明。
Balances Hi-C contact matrices (ICE via cooler.balance_cooler, KR/SCALE/VC context), computes distance-decay expected with cooltools (expected_cis per-diagonal P(s), expected_trans scalar), builds observed/expected (O/E) matrices, and diagnoses polymer state from the P(s) log-derivative. Covers the within-matrix-vs-cross-sample distinction (balancing is NOT a normalizer), the equal-visibility assumption that CNV/aneuploidy violates (use raw counts for copy-number), cis-only balancing, mad_max/blacklist masking before balancing, multiplicative cooler weights vs divisive juicer weights, and the resolution-vs-depth budget. Use when balancing a .cool/.mcool, computing expected or P(s), making O/E matrices for compartments/loops, deciding ICE vs KR vs SCALE, choosing a resolution for a given depth, or troubleshooting NaN/all-NaN balanced matrices; route cross-sample comparison to hic-differential.
中文简介正在准备中。当前显示原文说明。
Bulk-query Ensembl BioMart (and other BioMart instances) for cross-database ID mapping, gene/transcript/exon coordinates, and ortholog tables. Use when batch-converting Ensembl IDs to other namespaces (HGNC, RefSeq, UniProt, Entrez), pulling gene coordinate tables for thousands of genes, building ortholog wide-tables across species, or replacing slow Ensembl REST loops with one-shot bulk export. Encodes BioMart's XML query format, R biomaRt vs Python pybiomart trade-off, mart-vs-dataset hierarchy, and the URL endpoint that's BioMart-specific (separate from rest.ensembl.org).
中文简介正在准备中。当前显示原文说明。
Run build, deploy, inspect, and fix loops for .NET MAUI apps that already have MAUI DevFlow integrated. USE FOR: launching MAUI apps, selecting devices or emulators, waiting for or recovering agent connections, broker/port/adb connectivity issues, visual tree inspection, screenshots, UI interaction, Blazor WebView CDP debugging, reading DevFlow logs, and iterative app debugging. DO NOT USE FOR: first-time DevFlow package setup (use maui-devflow-onboard), or generic desktop automation unrelated to MAUI. INVOKES: maui devflow CLI, dotnet CLI, Android adb/android tools, and Apple simctl tools.
中文简介正在准备中。当前显示原文说明。
Use when a dispatcher, operations manager, or owner asks how to plan loads to minimize empty (deadhead) miles, sequence freight efficiently, balance regional capacity, design dedicated loops, or build a load-pairing strategy across customers. Covers loaded-ratio math, lane balance, region-pair pricing, and software-supported optimization.
中文简介正在准备中。当前显示原文说明。
Analyze undirected cycle structure within a valid DAG using the source paper's orientation and contraction method. Use for cycle-basis analysis with explicit graph and reduction choices. NOT for directed feedback loops, workflow deadlocks, or runtime resilience claims.
中文简介正在准备中。当前显示原文说明。
Diagnose when intuitive judgment, agent confidence, or expert routing can be trusted by classifying environment validity, feedback quality, and task-boundary fit. Use for confidence calibration, agent routing, expertise audits, and escalation design. NOT for deterministic implementation tasks, pure syntax debugging, or domains with explicit verifiable answers.
中文简介正在准备中。当前显示原文说明。
Diagnose when intuitive judgment, agent confidence, or expert routing can be trusted by classifying environment validity, feedback quality, and task-boundary fit. Use for confidence calibration, agent routing, expertise audits, and escalation design. NOT for deterministic implementation tasks, pure syntax debugging, or domains with explicit verifiable answers.
中文简介正在准备中。当前显示原文说明。