1.4K skills · sorted by relevance
Diagnose unnecessary rebuilds in MSBuild by examining build logs, file timestamps, and declared inputs and outputs. Identify why work repeats, guide fixes when requested, and verify that a second unchanged build skips the expected targets.
Design workflows that evaluate and refine AI outputs through self-review, separate generation and evaluation roles, tests, or weighted scoring criteria. Includes patterns for structured feedback, iteration limits, and stopping when quality stops improving.
Save agent memory and outputs in Cua Volume, shared storage that persists beyond a Space, the working environment. Hand files to the user, read shared references, and check synchronization, version history, and conflicting edits.
Prepare a ChatGPT Apps submission JSON file by inspecting an MCP server, which exposes tools to AI agents. Generate app details, tool behavior justifications, and test cases, then report review findings and missing output schemas.
Explain how input features contribute to machine-learning predictions using SHAP. Select an explanation method and reference data, validate contributions against model outputs, and create charts for individual predictions or patterns across samples, including models with multiple outputs.
Investigate failed Power Automate cloud flows through FlowStudio MCP. Inspect each action’s inputs and outputs, trace incorrect data upstream, fix expression or data issues, and replay a previous run to check the repair.
Add error handling to n8n workflows with connected error outputs, retries, and failure notifications. Return HTTP status codes that explain what went wrong, distinguish invalid requests from service failures, and keep internal error details out of public responses.
Validate a KrillinAI production plan combining subtitles, dubbing, landscape or portrait videos, and covers. The skill explains how to run stages separately, verify their files, and resume failed work using existing outputs.
Track tasks that require several outputs by listing every deliverable, updating a visible checklist, and verifying that each result exists. Helps prevent work from ending while requested files, reports, or other outputs remain unfinished.
Generate images, posters, and illustrations from written descriptions using an external image API, with Qwen-Image as the default. Save the finished image in the agent's working environment and display it through the artifact presentation tool.
Use Jev to choose among available tools, skills, or agents, or assess outputs against explicit criteria. Receive structured selections, scores, or probabilities to support decisions within a defined set of options.
Build failure handling into n8n workflows with separate error paths, retries, alerts, and clear API responses. Use it for webhooks and unattended jobs to make failures visible and return status codes that explain what went wrong.
Write and repair custom MSBuild targets, the steps that organize build tasks. Preserve existing build sequences, choose appropriate insertion points, distinguish returned results from rebuild checks, and register generated files for cleanup.
Turn repeatable outputs of a one-person company into compounding assets. Use when Codex needs to explain asset-compounding concepts when needed, verify prerequisite outputs, ask one question at a time, present multiple assetization priorities, and write user-confirmed outputs into `opc-doc/`.
The English summary is being prepared. Showing the original description.
Validates agent outputs against expected schemas and quality criteria. Ensures outputs meet structural requirements and content standards. Activate on 'validate output', 'output validation', 'schema validation', 'check output', 'output quality'. NOT for confidence scoring (use dag-confidence-scorer) or hallucination detection (use dag-hallucination-detector).
The English summary is being prepared. Showing the original description.
Score LLM and agent outputs using LLM-as-judge techniques — direct scoring against rubrics or pairwise comparison between two outputs. Includes built-in bias mitigation for position bias, length bias, and self-enhancement bias. Load when the user asks to score an output, judge a response, evaluate against a rubric, compare two outputs, do direct scoring, run pairwise comparison, or says "rate this", "which response is better", "score this against the rubric", "judge this output", "LLM as judge this". Sub-skill of eval-output orchestrator.
The English summary is being prepared. Showing the original description.
Check code and agent outputs, track reliability scores, and roll back changes that fail a chosen threshold. Includes quality reports, live monitoring, and integration with automated build and deployment workflows.
Guide market prediction workflows with LSTM, Transformer, and N-BEATS models using neural-trader commands and result storage. It explicitly identifies version 2.8.11 outputs as placeholders, so they cannot be treated as fitted models or evidence of prediction accuracy.
Turn an approved specification into pseudocode, a plain outline of program logic, and an architecture defining module responsibilities and API inputs and outputs. After user confirmation, proceed with implementation and unit tests based on those designs.
Check untrusted text before passing it to a model or executing it. Using AIDefence, it provides safety checks, detailed threat classifications, confidence scores, and scanning for attempts to override instructions or extract data.
Design how an AI agent trims long conversations and large tool outputs while retaining decisions, constraints, failed approaches, and sources. Compare summarization strategies, choose compression triggers, and consider separate subagents to keep intermediate information out of the main conversation.
Evaluate BMad skills using your agent's configured runner. Compare results with the model alone or another version, grade outputs against criteria, and test whether descriptions trigger correctly, preserving transcripts and evidence for each run.
Optimize scientific Python workloads for NVIDIA GPUs by choosing suitable libraries and reducing bottlenecks. Compare outputs with the CPU version and measure complete processing time, including data transfers, to determine whether the GPU implementation provides a useful improvement.
Helps organize biological datasets and models in LaminDB. Covers data registration, table validation, annotation with standardized biological terms, metadata searches, and tracking analysis inputs and outputs so researchers can preserve the context needed to repeat their work.