TalonT-Org
- 129 skills
- 0 followers
- 1 week ago last updated
- ▌ Exp Lens Estimand Clarity · talont-orgCreate Estimand Clarity experimental design analysis decomposing the implicit estimand from code vs. explicit claims from prose. Evidential lens answering "What exactly is the claim?"
- ▌ Exp Lens Severity Testing · talont-orgAnalyze severity of experimental tests — adversarial cases, negative controls, falsification tests, easy-pass detection, and confirmatory theater. Falsificationist lens answering "Would this design have caught the error?"
- ▌ Exp Lens Validity Threats · talont-orgCreate a validity threat matrix identifying alternative explanations and design mitigations. Adversarial lens answering "What alternative explanations survive?"
- ▌ Validate Review Decisions · talont-orgValidate review-decisions audit findings with mandatory intent analysis and seven evidence-gathering rules. Adds docstring-as-contract recognition, deliberate-change detection, test-as-intent-signal, consumer-impact verification, architectural feasibility checks, behavioral simulation, and symmetry-as-design recognition to the standard validation workflow. Use when validating reports from audit-review-decisions specifically.
- ▌ Arch Lens Error Resilience · talont-orgCreate Error/Resilience architecture diagram showing failure handling, recovery mechanisms, and circuit breakers. Diagnostic lens answering "How are failures handled?"
- ▌ Exp Lens Unit Interference · talont-orgCreate Unit Interference experimental design diagram showing unit hierarchy, cluster structure, shared resources, and SUTVA violation pathways. Causal-Structural lens answering "What is the unit, and can treatments spill over?"
- ▌ Planner Refine Assignments · talont-orgRefine elaborated assignments for a single phase via parallel L0 subagents (L1+L0 pattern), using per-phase context file with peer_summaries for cross-phase visibility
- ▌ Vis Lens Methodology Norms · talont-orgCreate Methodology Norms visualization planning spec showing ML sub-area mandatory figures, community conventions, and coverage gaps. Methodology-Normative lens answering "Which domain-specific figures are expected by reviewers?"
- ▌ Arch Lens Module Dependency · talont-orgCreate Module Dependency architecture diagram showing package coupling, layering, and fan-in/fan-out. Structural lens answering "How are modules coupled?"
- ▌ Arch Lens Repository Access · talont-orgCreate Repository/Data Access architecture diagram showing the repository pattern, entity relationships, and data access patterns. Data-centric lens answering "How is data accessed?"
- ▌ Exp Lens Causal Assumptions · talont-orgCreate Causal Assumptions experimental design diagram showing confounders, mediators, colliders, and identification strategy. Causal-structural lens answering "What causal assumptions support this design?"
- ▌ Exp Lens Iterative Learning · talont-orgCreate Iterative Learning experimental design diagram showing factor space exploration, adaptive allocation, and next-experiment recommendations. Decision-Theoretic lens answering "How does this maximize learning per cost?"
- ▌ Exp Lens Pipeline Integrity · talont-orgCreate Pipeline Integrity experimental design diagram showing data splits, leakage points, preprocessing order, and label contamination. Integrity lens answering "Could data handling create optimistic bias?"
- ▌ Exp Lens Variance Stability · talont-orgCreate a variance analysis profile assessing whether signals exceed noise and whether results are stable across random seeds. Stability lens answering "Is the signal larger than the noise?"
- ▌ Implement Worktree No Merge · talont-orgImplementation executor. ALWAYS invoke this skill when instructed to implement a plan in a worktree. Do not read the plan or edit files directly — use this skill first to load the full implementation workflow.
- ▌ Exp Lens Measurement Validity · talont-orgAnalyze measurement validity for experimental design — auditing metric-construct alignment, proxy validity, reliability, sensitivity, and consequential validity. Argumentative lens answering "Do measurements justify the interpretation?"
- ▌ Planner Elaborate Assignments · talont-orgElaborate all assignments for a target phase via parallel L0 subagents. One invocation per phase; spawns one L0 per assignment concurrently. (Pass 2 loop body)
- ▌ Planner Assess Review Approach · talont-orgAssess each work package for review-approach benefit before implementation. Writes review_approach_assessment.json; does NOT invoke review-approach.
- ▌ Exp Lens Randomization Blocking · talont-orgCreate Randomization & Blocking experimental design diagram showing assignment mechanisms, blocking factors, and comparability sources. Design-Structural lens answering "Where does comparability come from?"
- ▌ Exp Lens Sensitivity Robustness · talont-orgCreate Sensitivity & Robustness experimental design analysis identifying load-bearing analytic choices and untested perturbations. Robustness lens answering "Which assumptions are load-bearing?"
- ▌ Planner Validate Task Alignment · talont-orgValidate that plan phases and WPs align with the stated task
- ▌ Exp Lens Comparator Construction · talont-orgCreate Comparator Construction experimental design analysis assessing whether baselines and controls are fair and relevant. Counterfactual lens answering "Is the comparator fair and relevant?"
- ▌ Exp Lens Exploratory Confirmatory · talont-orgAssess whether analytic decisions were pre-specified or post-hoc and whether exploratory/confirmatory norms are aligned. Boundary lens answering "Is this discovery or test, and are norms aligned?"
- ▌ Exp Lens Reproducibility Artifacts · talont-orgCreate Reproducibility Artifacts experimental design diagram showing run instructions, environment capture, data availability, determinism controls, and audit trail. Transparency lens answering "Could an independent party reproduce this?"
- ▌ Exp Lens Benchmark Representativeness · talont-orgCreate Benchmark Representativeness experimental design diagram showing coverage matrix, generalization gaps, and untested regions. Generalizability lens answering "Does this generalize beyond the test bed?"
- ▌ Resolve Design Review · talont-orgTriage STOP verdict findings from review-design, classifying each as ADDRESSABLE/STRUCTURAL/DISCUSS using parallel subagents. If any are ADDRESSABLE or DISCUSS, generate revision_guidance and emit resolution=revised. If all are STRUCTURAL, emit resolution=failed for terminal stop.
- ▌ Vis Lens Chart Select · talont-orgCreate Chart Type Selection visualization planning spec showing encoding channel assignments, Cleveland-McGill perceptual hierarchy, and data-type→chart-type matrix. Typological lens answering "Which chart type is perceptually optimal for this data?"
- ▌ Vis Lens Color Access · talont-orgCreate Chromatic Accessibility visualization planning spec showing colorblind safety (Okabe-Ito, Paul Tol palettes), perceptual uniformity checks (viridis/cividis pass; jet/rainbow fail), non-color redundant encoding (shape + line-style), and text size minimums. Chromatic lens answering "Is the color encoding accessible and perceptually uniform?"
- ▌ Vis Lens Figure Table · talont-orgCreate Decisional Layout visualization planning spec showing figure-vs-table selection heuristics: tables win for exact values, ≤5 items, leaderboards, and ablation matrices; figures win for trends, distributions, and spatial patterns; borderline cases recommend both. Decisional lens answering "Should this result be a figure or a table?"
- ▌ Arch Lens C4 Container · talont-orgCreate C4 Container architecture diagram showing static structure, building blocks, and technology choices. Anatomical lens answering "How is it built?"
- ▌ Arch Lens Data Lineage · talont-orgCreate Data Lineage architecture diagram showing information flow, transformations, and storage destinations. Data-centric lens answering "Where is the data?"
- ▌ Arch Lens Process Flow · talont-orgCreate Process/Execution Flow architecture diagram showing runtime behavior, state transitions, and decision points. Physiological lens answering "How does it behave?"
- ▌ Audit Review Decisions · talont-orgAudit merged PR review threads for agreed-but-deferred suggestions (design decisions, future work, out-of-scope items) that were never implemented. Mines REVIEW-FLAG markers from resolve-review and legacy keyword signals. Produces a structured markdown report with VALID/RESOLVED/STALE classifications and annotates processed threads with [AUDIT] markers to prevent re-identification on future runs.
- ▌ Planner Extract Domain · talont-orgExtract domain knowledge and naming conventions for planning context
- ▌ Planner Reconcile Deps · talont-orgPost-Pass-3 global dependency DAG construction — detects implicit dependencies individual WP sessions could not see
- ▌ Vis Lens Caption Annot · talont-orgCreate Annotative Caption visualization planning spec showing declarative titles, axis labels with units, error definition in legend, baseline references, sample sizes, and venue-specific caption format. Annotative lens answering "Are figure captions and axis labels fully self-contained?"
- ▌ Vis Lens Multi Compare · talont-orgCreate Compositional Layout visualization planning spec showing small-multiples vs overlay decisions, faceting strategy (row/col), shared-axis alignment, grouped vs stacked bars, factorial interaction plots, and panel reading order. Compositional lens answering "Which layout best reveals the comparison structure?"
- ▌ Audit Defense Standards · talont-orgAudit the codebase against defense standards derived from historical bug patterns. Standards accumulate over time as new patterns are discovered via audit-bugs and design-guards. Use when user says "audit defenses", "audit defense standards", "check defenses", or "defense audit".
- ▌ Planner Consolidate Wps · talont-orgAnalyze WP complexity per phase and emit consolidation group manifests for trivial WP merging (L1+L0 pattern)
- ▌ Planner Elaborate Phase · talont-orgElaborate a single phase into a full result, parallel-safe — receives plan snapshot + target phase ID
- ▌ Planner Generate Phases · talont-orgGenerate high-level phases from project analysis (Pass 1 entry point)
- ▌ Resolve Merge Conflicts · talont-orgGoal-aware resolution of rebase conflicts when merging a conflict-resolution worktree back into the integration branch. Analyzes the intent of each side of a conflict, resolves it in-place when confidence is HIGH or MEDIUM, and escalates when LOW.
- ▌ Resolve Research Review · talont-orgFetch PR review comments from review-research-pr, run research-aware intent validation (ACCEPT/REJECT/DISCUSS), apply targeted fixes, escalate unrerunnable findings, and post inline replies. Exit 0 drives recipe re_push_research; exit non-zero halts the cycle.
- ▌ Exp Lens Fair Comparison · talont-orgCreate a comparison fairness matrix assessing whether alternatives are evaluated under symmetric constraints. Fairness lens answering "Are alternatives compared under symmetric constraints?"
- ▌ Exp Lens Governance Risk · talont-orgCreate a risk register and stakeholder impact assessment for experiments with deployment implications. Governance lens answering "What risks arise from acting on this result?"
- ▌ Vis Lens Reproducibility · talont-orgCreate Replicative Reproducibility visualization planning spec showing data availability, preprocessing parameter disclosure (bin widths, smoothing windows), plotting library/version, random seeds, and code reference per figure. Replicative lens answering "Can the figures be reproduced from the data and code?"
- ▌ Arch Lens State Lifecycle · talont-orgCreate State Lifecycle architecture diagram showing field contracts, validation gates, and resume safety. Contract overlay lens answering "How is state corruption prevented?"
- ▌ Vis Lens Story Arc · talont-orgCreate Narrative Story Arc visualization planning spec showing visual consistency across the report (same color = same model everywhere), logical figure progression, redundant figure detection, and narrative dependency between figures. Narrative lens answering "Do the figures tell a coherent story across the report?"
- ▌ Arch Lens Scenarios · talont-orgCreate Scenarios architecture diagram showing end-to-end user journeys and component cooperation validation. Validation lens answering "Do the components work together?"
- ▌ Audit Feature Gates · talont-orgAudit feature flag isolation — traces import chains, runtime gates, tool/skill tag coverage, UI surfaces, and test markers to detect leakage and miswiring.
- ▌ Build Execution Map · talont-orgAnalyze issue dependencies and produce a dispatch execution map for parallel orchestration
- ▌ Compose Research Pr · talont-orgReads a PR prep file and validated experiment diagrams, composes the PR body, and creates the GitHub PR. Part 3 of 3 in the decomposed research-PR flow.
- ▌ Open Integration Pr · talont-orgCreate an integration PR for the merge-prs. Reads pr_order_file JSON, generates a rich PR body with per-PR details, arch-lens diagrams, and carried-forward Closes #N references. Closes all collapsed PRs with a comment after creation. Use inside the merge-prs after all PRs have been merged into the integration branch.
- ▌ Prepare Research Pr · talont-orgReads a research report and experiment plan, synthesizes a recommendation, selects 1-2 exp-lens lenses, writes a context file per lens, and writes a PR prep file. Does NOT open a PR. Part 1 of 3 in the decomposed research-PR flow.
- ▌ Validate Test Audit · talont-orgValidate test audit findings with test-domain semantic rules and intent analysis. Adds import-path-as-contract recognition, precondition-as-assertion detection, provenance verification, split-era lifecycle awareness, and deletion-vs-improvement distinction to the standard validation workflow. Use when validating reports from audit-tests specifically.
- ▌ Arch Lens Deployment · talont-orgCreate Deployment/Physical architecture diagram showing infrastructure topology, process boundaries, and network communication. Physical lens answering "Where does it run?"
- ▌ Implement Experiment · talont-orgDeploy experiment artifacts in an isolated git worktree following an approved experiment plan, with per-phase commits.
- ▌ Make Experiment Diag · talont-orgInteractive selection of experimental design lens for visualizing experiment methodology. Routes to the appropriate exp-lens-* skill.
- ▌ Vis Lens Antipattern · talont-orgCreate Anti-Pattern Detection visualization audit showing severity-tiered catalog of visualization anti-patterns present in or planned for the experiment. Diagnostic lens answering "Which visualization anti-patterns are present?"
- ▌ Vis Lens Uncertainty · talont-orgCreate Uncertainty Representation visualization planning spec showing error bar definitions, distribution-aware alternatives, and multi-seed variance protocols. Statistical lens answering "How is uncertainty honestly represented?"
- ▌ Arch Lens Concurrency · talont-orgCreate Concurrency architecture diagram showing parallel execution patterns, thread pools, synchronization, and barriers. Physiological lens answering "How does parallelism work?"
- ▌ Arch Lens Development · talont-orgCreate Development architecture diagram showing project structure, build tools, and quality gates. Development lens answering "How is it built and tested?"
- ▌ Arch Lens Operational · talont-orgCreate Operational architecture diagram showing CLI workflows, configuration, and observability. Administration lens answering "How is it run and monitored?"
- ▌ Exp Lens Error Budget · talont-orgAnalyze statistical error budget showing Type I/II errors, power, minimum detectable effect, multiplicity corrections, and sequential monitoring. Statistical lens answering "Are error risks sized and controlled?"
- ▌ Planner Elaborate Wps · talont-orgElaborate all work packages for a target phase via parallel L0 subagents. One invocation per phase; spawns one L0 per WP concurrently. (Pass 3 loop body)
- ▌ Planner Refine Phases · talont-orgRefine elaborated phases with cross-phase visibility via parallel L0 subagents (L1+L0 pattern)
- ▌ Resolve Claims Review · talont-orgFetch claim findings from audit-claims, run citation-aware intent validation (ACCEPT/REJECT/DISCUSS), apply targeted citation fixes, escalate findings requiring experiment reruns, and post inline replies.
- ▌ Retry Worktree · talont-orgWorktree retry executor. ALWAYS invoke this skill when instructed to continue or retry an implementation in an existing worktree. Do not resume editing files directly — use this skill first to load the retry workflow.
- ▌ Run Experiment · talont-orgExecute a designed experiment in a worktree and collect structured results. Supports --adjust retry mode.
- ▌ Collapse Issues · talont-orgIdentify clusters of related triaged GitHub issues sharing the same recipe route and collapse them into a single combined issue with full content from all originals. Closes originals with cross-reference comments. Inverse of issue-splitter.
- ▌ Dry Walkthrough · talont-orgPlan validation executor. ALWAYS invoke this skill when instructed to validate or dry-walkthrough a plan. Do not read the plan or trace changes directly — use this skill first to load the validation workflow.
- ▌ Elaborate Phase · talont-orgElaborate a migration plan phase into a complete self-contained implementation plan. Use when user says "elaborate phase", "elaborate phase N", or "phase elaboration". Assesses codebase, writes detailed phase plan, then validates with dry walkthrough.
- ▌ Generate Report · talont-orgSynthesize experiment results into a structured research report in the research/ folder. Supports --inconclusive flag.
- ▌ Migrate Recipes · talont-orgApply versioned migration notes to an AutoSkillit recipe. Use when user confirms migration, called by agent or autoskillit migrate CLI, or invoked directly.
- ▌ Plan Experiment · talont-orgConvert a scope report into a structured experiment plan with hypothesis, variables, phases, and success criteria.
- ▌
- ▌ Review Approach · talont-orgResearch modern solutions and approaches for issues or features proposed in a report or plan. Use when user says "review approach", "review approaches", "research solutions", or wants external validation of a proposed direction.
- ▌ Pipeline Summary · talont-orgCreate a GitHub issue and PR summarizing pipeline bugs and fixes. Use when a pipeline run completes with accumulated bug fixes on a feature branch.
- ▌ Resolve Failures · talont-orgFailure resolution executor. ALWAYS invoke this skill when instructed to fix test failures in a worktree. Do not read test output or edit code directly — use this skill first to load the failure resolution workflow.
- ▌ Setup Environment · talont-orgPre-flight environment gate for the research recipe. Reads the experiment plan, detects the required environment type, builds a Docker image or creates a host micromamba environment, and emits an env_mode verdict consumed by downstream steps.
- ▌ Vis Lens Temporal · talont-orgCreate Temporal Dynamics visualization planning spec showing axis scaling (linear vs log), smoothing disclosure, epoch/step alignment, run aggregation (mean + variance bands), early-stopping markers, and wall-clock vs step-count x-axis. Temporal lens answering "Are training dynamics shown clearly and honestly?"
- ▌ Arch Lens Security · talont-orgCreate Security architecture diagram showing trust boundaries, validation layers, and process isolation. Security lens answering "Where are the trust boundaries?"
- ▌ Implement Worktree · talont-orgWorktree implementation executor. ALWAYS invoke this skill when instructed to implement a plan in a worktree with testing and merging. Do not read the plan or edit files directly — use this skill first to load the full implementation workflow.
- ▌ Plan Visualization · talont-orgOrchestrates 2–4 vis-lens skills in parallel to produce a figure inventory (visualization-plan.md) and a report-placement outline (report-plan.md). Runs after design review GO, before worktree creation.
- ▌ Planner Refine Wps · talont-orgRefine elaborated work packages with cross-phase visibility via per-phase L0 subagents (L1+L0 pattern)
- ▌ Review Research Pr · talont-orgAutomated diff-scoped research PR review using parallel audit subagents aligned to research quality dimensions. Posts inline GitHub review comments and submits a summary verdict. Use after a research PR is opened to gate on review approval.
- ▌ Vis Lens Always On · talont-orgCreate Always-On visualization triage report running three sequential analysis passes (anti-pattern, accessibility, annotation completeness) and emitting a combined PASS|WARN_N|FAIL_N verdict. Composite lens answering "What are the blocking visualization issues?"
- ▌ Diagnose CI · talont-orgDiagnostic executor for CI failures. ALWAYS invoke this skill when instructed to diagnose CI failures. Do not fetch CI logs directly — use this skill first to load the diagnosis workflow.
- ▌ Investigate · talont-orgDeep investigation of errors, bugs, or codebase questions without making any code changes. Use when user mentions investigate, understand, explore, analyze, or pastes error tracebacks. Spawns parallel subagents for comprehensive exploration.
- ▌ Make Groups · talont-orgBreak a large plan, architecture proposal, or feature document into sequenced implementation groups for the make-plan pipeline. Use when user says "make groups", "group requirements", "sequence groups", or wants to decompose a large document into ordered implementation units.
- ▌ Verify Diag · talont-orgVerify an architecture diagram against the actual codebase. Checks component existence, connection accuracy, and read/write directionality. Use when user says "verify diagram", "verify diag", "check diagram", or wants to validate diagram accuracy.
- ▌ Audit Claims · talont-orgParallel subagent-driven claim extraction and citation integrity audit for research PRs. Extracts claims by section, matches against available evidence, classifies unsupported claims as findings, and emits a verdict for recipe routing.
- ▌ Write Recipe · talont-orgGenerate YAML recipes for .autoskillit/recipes/. Use when user says "make script skill", "generate script", "script a workflow", "write a script", "create a script", "new recipe", "write a pipeline", or when loaded by other skills for script formatting.
- ▌ Design Guards · talont-orgInvestigate a bug pattern audit report and design architectural guards (tests, contracts, structural changes) that provide immunity to each identified pattern. Use when user says "design guards", "design defenses", or wants architectural solutions for bug patterns.
- ▌ Enrich Issues · talont-orgBackfill structured requirements on existing GitHub issues triaged with recipe:implementation labels. Scans candidates, skips already-enriched issues, performs codebase-grounded analysis, and appends a Requirements section in REQ-{GRP}-NNN format via gh issue edit.
- ▌ Make Campaign · talont-orgInteractively author a campaign recipe YAML through a 6-phase guided workflow. Use when user says "make campaign", "create campaign", "author campaign", "new campaign recipe", or wants to decompose a campaign goal into dispatches.
- ▌ Prepare Issue · talont-orgCreate a single GitHub issue and immediately triage it — dedup check, classification (recipe:implementation or recipe:remediation), mixed-concern detection, and label application. Use when user says "open an issue", "create an issue", "file an issue", or "file a bug". The user-facing counterpart to report_bug.
- ▌ Review Design · talont-orgValidate an experiment plan before execution using triage-first, fail-fast dimensional analysis with an adversarial red-team. Emits verdict (GO/REVISE/STOP), experiment_type, evaluation_dashboard, and revision_guidance.
- ▌ Setup Project · talont-orgExplore a target project and generate tailored recipes and config through an interactive workflow. Use when user wants to onboard a new project to AutoSkillit, says "setup project", or wants a starting point config.
- ▌ Triage Issues · talont-orgAnalyze open GitHub issues and produce a sequenced implementation plan — grouping issues into parallel batches, ordering those batches, and tagging each issue with its recipe route. Use when user says "triage issues", "prioritize issues", or "plan issue order".