543 skills · sorted by relevance
Capture confidence ratings before and after a learning attempt to identify overconfidence and underconfidence patterns. Use when a student wants to understand how well they actually know something versus how well they think they know it.
The English summary is being prepared. Showing the original description.
Capture confidence ratings before and after a learning attempt to identify overconfidence and underconfidence patterns. Use when a student wants to understand how well they actually know something versus how well they think they know it.
The English summary is being prepared. Showing the original description.
Assigns confidence scores to agent outputs based on multiple factors including source quality, consistency, and reasoning depth. Produces calibrated confidence estimates. Activate on 'confidence score', 'how confident', 'certainty level', 'output confidence', 'reliability score'. NOT for validation (use dag-output-validator) or hallucination detection (use dag-hallucination-detector).
The English summary is being prepared. Showing the original description.
Use when assigning a confidence level to an analytical judgment, the user asks "how confident are we?" / "what is the confidence on X?", or the orchestrator's tradecraft pipeline calls for a confidence level before publishing. Provides three named levels (High, Moderate, Low) with 0-100 score ranges aligned to STIX, and the rule that confidence cannot exceed what the weakest load-bearing claim supports.
The English summary is being prepared. Showing the original description.
Generate candidate positions and orientations for small molecules binding to proteins with DiffDock and DiffDock-L. Prepare single or batch docking runs, organize predicted poses by model confidence, and review validation needs without treating confidence as binding affinity.
Compare evidence from cyber operations to assess which attacker may be responsible. Weigh shared infrastructure, attack methods, malware similarities, timing, and language clues against competing explanations, then report attribution findings with justified confidence levels.
Reads model uncertainty from token log-probabilities. Covers collecting logprobs and aggregating multi-token labels, confidence tiers and margins for triage, calibration assessment with ECE, Brier scores, and reliability diagrams against human labels, responsible downstream use of confidence, and reproducibility archives. Use when the user asks about classifier confidence by item, logprobs, top-k tokens, calibration, ECE, Brier, reliability diagrams, or routing low-confidence cases to human review. Codebook and validation design go to text-classification.
The English summary is being prepared. Showing the original description.
VRChat エコシステムツール（ModularAvatar / liltoon / VRCFury / AvatarOptimizer 等）の ナレッジを 4 フェーズで段階的に調査・検証し `knowledge/*.md` に再現可能な形で蓄積する。 ソースコード読解 → inspect 実測まで進めると confidence を `high` に昇格できる。 トリガー: 新しい VRChat コミュニティツールに初めて遭遇した、 既存ナレッジの confidence が low で作業に支障がある、 ユーザーから「このツールについて調べて」と依頼された場合。
The English summary is being prepared. Showing the original description.
Design text extraction for forms, invoices, and quizzes using regex, rules that match text patterns. Check extraction confidence and send uncertain items to an LLM for review, with guidance on testing and tracking processing costs.
Assess a repository without changing it and report its type, suggested agent roles, risk, MCP connections, test confidence, and publication readiness. Use the results for setup decisions, comparisons over time, or automated risk checks in CI.
Build an OpenCTI pipeline that adds external context to indicators of compromise, such as suspicious IP addresses. Connect intelligence services, relate indicators to known attackers and campaigns, and score confidence to help analysts prioritize investigations.
Plan user research by matching methods to study goals and explaining participant numbers and confidence. Synthesize tagged observations into recurring findings, distinguish isolated anecdotes, and organize evidence so research results can be found and reused.
Find people and assess identity matches using public information from social media, official records, and reverse lookups. The skill cross-checks independent sources, reports confidence levels, and requires at least three matching identifiers for a high-confidence result.
Build probability estimates from historical frequencies and supporting or opposing evidence. Structure forecasts with explicit assumptions and outcome criteria, update them as evidence changes, and assess past predictions using accuracy and confidence calibration.
Opt-in evidence-first causal diagnosis for bugs, browser or app/device failures, flaky behavior, and performance regressions. Activate only when the user explicitly invokes `$diagnose` or explicitly asks to use the named diagnose skill. Do not activate merely because the user mentions a bug, asks why something failed, requests debugging or a fix, or describes unexpected behavior. Once explicitly invoked, locate likely causes, instrument relevant boundaries with extensive structured logging, reproduce the issue, analyze the collected logs, rank causes with confidence scores, and pursue 100% operational confidence while probes can increase confidence.
The English summary is being prepared. Showing the original description.
Routes, triages, flags and rates a piece of text with a probability for every option: which department or queue a ticket goes to, which intent a message expresses, whether a yes/no condition holds (is this tool call grounded, should the agent ask first, is the customer angry), and how severe or urgent it is on a rubric, each with a calibrated confidence. Runs Claude Haiku in the request and response shapes of OpenAI's Decisions API, for when no purpose-built decision model is at hand. Use for "route this ticket", "triage these", "classify with a confidence", "give me a probability that", "rate the severity", "score against this rubric", guardrail checks before a tool call, or "a Decisions API without OpenAI".
The English summary is being prepared. Showing the original description.
Reads a model's own uncertainty from token log-probabilities: collecting and aggregating label logprobs, confidence tiers and margins for triage, calibration against human labels (ECE, Brier, reliability diagrams), downstream use, and reproducibility. Use for per-item classifier confidence or routing low-confidence cases to human review; codebook and validation design go to text-classification.
The English summary is being prepared. Showing the original description.
DiffDock and DiffDock-L diffusion-based molecular docking. Use for blind protein-small-molecule pose prediction from a PDB file or sequence plus SMILES/SDF/MOL2, batch docking over a CSV of complexes, virtual screening triage, sampling multiple poses per complex, and reading the confidence score correctly. Also trigger on DiffDock, DiffDock-L, inference.py, confidence_model, samples_per_complex, ESM embedding preparation for docking, or blind docking without a defined box. Not for binding affinity prediction — the confidence score ranks pose plausibility, not potency.
The English summary is being prepared. Showing the original description.
Identify insects using body plan analysis, dichotomous keys to order, wing venation, mouthpart type, antennae form, leg and tarsal structure, and confidence levels. Covers the fundamental hexapod body plan verification, a simplified dichotomous key to major orders, wing venation and type analysis, mouthpart classification, antennae morphology, leg specialization and tarsal formula, and a structured confidence assessment framework. Use when you need to identify an unknown insect beyond preliminary order placement, are working through a specimen for taxonomic study, want to distinguish between similar orders or families, or need to assign a confidence level to a field identification.
The English summary is being prepared. Showing the original description.
Epistemic transparency — acknowledging uncertainty, flagging limitations, avoiding overconfidence, and communicating what is known, unknown, and uncertain with proportional confidence. Maps the HEXACO personality dimension to AI reasoning: truthful calibration of confidence, proactive disclosure of gaps, and resistance to the temptation to appear more certain than warranted. Use before presenting a conclusion, when answering questions where knowledge is partial or inferred, after noticing a temptation to state uncertain information as certain, or when a user is making decisions based on provided information.
The English summary is being prepared. Showing the original description.
Add an external probabilistic classifier to a decision path without letting it take the path over. Covers finding the externally-graded rows you are already logging, measuring the confidence separation between the overrides it got right and the ones it got wrong, choosing an operating point your data licenses, failing open at the call site, and proving the whole arrangement offline with no API key. Applies to any service returning a score you can order — TypeSafe AI's System One models (Jev) are the worked example. Most of the method transfers to a local model or a second heuristic, with the parts that do not called out where they arise. Use when a rule-based path is wrong often enough to hurt, when someone proposes replacing a heuristic with a model, when a confidence threshold needs a defensible value, or when an oracle already in production has never been graded.
The English summary is being prepared. Showing the original description.
Records atomic learned behaviors with confidence scores. Project-scoped instincts are isolated per project and auto-promote to global scope at confidence threshold 0.8. Stores instincts in .claude/context/memory/instincts.jsonl
The English summary is being prepared. Showing the original description.
Validates agent outputs against schemas and quality criteria, scores confidence, detects hallucinations, monitors convergence, decides when to iterate, and synthesizes actionable feedback. Use when checking if a node's output is acceptable, scoring confidence, detecting fabricated content, deciding whether to re-execute, or generating improvement feedback. Activate on "validate output", "check quality", "confidence score", "hallucination check", "should we iterate", "improvement feedback". NOT for executing DAGs (use dag-runtime), planning DAGs (use dag-planner), or matching skills (use dag-skills-matcher).
The English summary is being prepared. Showing the original description.
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.
The English summary is being prepared. Showing the original description.