agent-almanac

<!-- AUTO:START:agents-intro -->

---

---

Panoramic synthesis through simultaneous multi-domain awareness — perceives the whole field before acting on any part, forming emergent insights from cross-domain resonances and tensions

---

---

APA 7th edition compliance specialist for academic manuscripts, covering citation formatting, table/figure styling, document structure, and Quarto/papaja implementation

---

---

---

---

Open-source contributor specializing in NVIDIA Claw ecosystem projects (OpenClaw, NemoClaw, NanoClaw) with security-aware audit, false positive prevention, and convention-strict PR workflow

CLI and terminal tool development specialist for Commander.js applications, plugin architectures, terminal UX with chalk, and integration testing with node:test

---

---

---

---

---

---

---

Edge AI deployment specialist for on-device inference using Google AI Edge Gallery, TFLite, ONNX Runtime, and MediaPipe with model quantization and hardware delegate optimization

Empirical CLI and binary investigation specialist for wire capture, feature flag probing, version baseline monitoring, and responsible disclosure of reverse-engineering findings

---

---

Open-source agent framework assessor that evaluates community health, supersession risk, architecture alignment, and governance sustainability to classify investment readiness before committing resources

---

---

---

---

---

---

---

---

---

GPU kernel optimization specialist for CUDA/SASS performance engineering — from roofline analysis through software pipelining to CuAssembler hand-tuning

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

Master skill composing the 5-step synoptic cycle for panoramic synthesis across multiple domains. Orchestrates meditate, expand-awareness, observe, awareness, integrate-gestalt, and express-insight into a coherent process that produces unified understanding rather than sequential compromise. Use when a problem genuinely spans 3+ domains and the interactions between domains matter more than depth in any one, when sequential analysis feels like compromise rather than integration, or before major architectural decisions affecting multiple stakeholders.

---

---

---

---

---

---

---

Systematically identify whether a GPU kernel is compute-bound, memory-bound, or latency-bound using roofline analysis, occupancy calculations, compute/load ratio per tile, and SASS instruction inspection. Produces a decision matrix for optimization strategy selection (cp.async, warp interleaving, tiling, double-buffering, or CuAssembler hand-tuning).

---

---

---

Analyze a tensegrity system by identifying compression struts and tension cables, classifying type (class 1/2, biological/architectural), computing prestress equilibrium, verifying stability via Maxwell's rigidity criterion, and mapping biological tensegrity (microtubules, actin, intermediate filaments). Use when evaluating tensegrity in architecture, robotics, cell biology, or any system with isolated compression in continuous tension.

---

---

---

---

Construct well-structured arguments using the hypothesis-argument-example triad. Covers formulating falsifiable hypotheses, building logical arguments (deductive, inductive, analogical, evidential), providing concrete examples, and steelmanning counterarguments. Use when writing or reviewing PR descriptions that propose technical changes, justifying design decisions in ADRs, constructing substantive code review feedback, or building a research argument or technical proposal.

---

---

---

---

---

---

---

---

Audit and repair Claude Code discovery symlinks for skills, agents, and teams. Compares registries against .claude/ directories at project and global levels, detects missing, broken, and extraneous symlinks, distinguishes almanac content from external projects, and optionally repairs gaps. Use after adding new skills or agents, after a repository rename or move, when slash commands stop working, or as a periodic health check.

Detect missing glyphs, icons, and HD variants by comparing registries against glyph mapping files, icon directories, and manifests. Reports gaps for skills, agents, and teams across all palettes.

---

---

---

---

---

---

Design and implement multi-stage CI/CD pipelines using GitHub Actions with matrix builds, dependency caching, artifact management, and secret handling. Create workflows that span linting, testing, building, and deployment stages with parallel execution and conditional logic. Use when setting up automated testing and deployment for a new project, migrating from Jenkins or CircleCI to GitHub Actions, implementing matrix builds across platforms, adding build caching, or creating multi-stage pipelines with security scanning and quality gates.

Build a plugin or adapter for a CLI tool using the abstract base class pattern. Covers defining the contract (static fields, required methods), choosing an installation strategy (symlink, copy, append-to-file), implementing detection, install/uninstall with idempotency, listing, auditing, and registering the plugin. Use when adding support for a new framework to a CLI installer, building a plugin system for any multi-target tool, or extending an existing adapter architecture.

---

---

---

Build a feature store using Feast for centralized feature management, configure offline and online stores for batch and real-time serving, define feature views with transformations, and implement point-in-time correct joins for ML pipelines. Use when managing features for multiple ML models, ensuring training-serving consistency, serving low-latency features for real-time inference, reusing feature definitions across projects, or building a feature catalog for discovery and governance.

Create production-ready Grafana dashboards with reusable panels, template variables, annotations, and provisioning for version-controlled dashboard deployment. Use when creating visual representations of Prometheus, Loki, or other data source metrics, building operational dashboards for SRE teams, migrating from manual dashboard creation to version-controlled provisioning, or establishing executive-level SLO compliance reporting.

---

---

---

---

---

---

---

---

---

Select a `delaySeconds` value when scheduling a loop wakeup via the `ScheduleWakeup` tool or the `/loop` slash command. Covers the three-tier cache-aware decision (cache-warm under 5 minutes, cache-miss 5 minutes to 1 hour, idle default 20 to 30 minutes), the 300-second anti-pattern, the [60, 3600] runtime clamp, the minute-boundary rounding quirk, and writing a telemetry-worthy `reason` field. Use when designing an autonomous loop, when a heartbeat cadence is being planned, when polling cadence is being tuned, or when post-hoc review of loop costs reveals interval mis-sizing.

---

---

---

---

---

---

Capture outbound HTTP and telemetry from a CLI harness at runtime. Covers capture-channel selection (transcript file vs verbose-fetch stderr vs outbound proxy vs on-disk state), hook-driven per-event capture vs long-running session capture, JSONL output format for diff-friendly artifacts, and the observability table that maps each target to the cheapest channel that captures it. Use when a static finding needs runtime confirmation, when a payload shape is needed for a client re-implementation, or when dark-vs-live disambiguation requires watching what the binary actually sends.

---

Conduct a blameless post-mortem analysis after an incident. Build timeline reconstruction, identify contributing factors, and generate actionable improvements. Focus on systemic issues rather than individual blame. Use after any production incident or service degradation, following a near-miss, when investigating recurring issues, or to share systemic learnings across teams.

---

Configure Prometheus Alertmanager with routing trees, receivers (Slack, PagerDuty, email), inhibition rules, silences, and notification templates for actionable incident alerting. Use when implementing proactive monitoring with automated incident detection, routing alerts to the appropriate team by severity, reducing alert fatigue through grouping and deduplication, integrating with on-call systems like PagerDuty, or migrating from legacy alerting to Prometheus-based alerting.

Deploy and configure an API gateway (Kong or Traefik) to handle API traffic management, authentication, rate limiting, request/response transformation, and routing. Covers plugin configuration, upstream services, consumer management, and integration with existing infrastructure. Use when multiple backend services need a unified API endpoint, when centralized authentication or rate limiting is required, when implementing API versioning, or when needing detailed analytics and load balancing for microservices.

---

Configure Kubernetes Ingress networking with NGINX Ingress Controller, cert-manager for automated TLS certificate management, path-based routing, rate limiting, and multi-domain hosting with SSL termination and load balancing. Use when exposing multiple Kubernetes services via a single load balancer, implementing path-based or host-based routing, automating TLS certificate issuance with Let's Encrypt, or setting up blue-green and canary deployments with traffic splitting.

Set up centralized log aggregation with Loki and Promtail (or ELK stack), including log parsing, label extraction, retention policies, and integration with metrics for correlation. Use when consolidating logs from multiple services into a searchable system, replacing local log files with centralized queryable storage, correlating logs with metrics and traces, implementing structured logging with label extraction, or troubleshooting production incidents requiring cross-service log analysis.

---

---

---

---

---

---

---

---

---

---

Unify metrics, logs, and traces for cohesive debugging. Implement exemplars for log-to-trace linking, build unified dashboards using RED/USE methods, and enable rapid root cause analysis across observability signals. Use when investigating complex incidents spanning multiple systems, reducing mean time to resolution, implementing distributed tracing, or moving from siloed tools to a unified observability platform.

---

---

Create a new agent definition file following the agent-almanac agent template and registry conventions. Covers persona design, tool selection, skill assignment, model choice, frontmatter schema, required sections, registry integration, and discovery symlink verification. Use when adding a new specialized agent to the library, defining a persona for a Claude Code subagent, or creating a domain-specific assistant with curated skills and tools.

---

---

---

Create R-based pictogram glyphs for skill, agent, or team icons in the visualization layer. Covers concept sketching, ggplot2 layer composition using the primitives library, color strategy, registration in the appropriate glyph mapping file and manifest, rendering via the build pipeline, and visual verification of the neon-glow output. Use when a new entity has been added and needs a visual icon for the force-graph visualization, an existing glyph needs replacement, or when batch-creating glyphs for a new domain.

---

---

---

---

---

Create a new SKILL.md file following the Agent Skills open standard (agentskills.io). Covers frontmatter schema, section structure, writing effective procedures with Expected/On failure pairs, validation checklists, cross-referencing, and registry integration. Use when codifying a repeatable procedure for agents, adding a new capability to the skills library, converting a guide or runbook into agent-consumable format, or standardizing a workflow across projects or teams.

---

Create a new team composition file following the agent-almanac team template and registry conventions. Covers team purpose definition, member selection, coordination pattern choice, task decomposition design, machine-readable configuration block, registry integration, and README automation. Use when defining a multi-agent workflow, composing agents for a complex review process, or creating a coordinated group for recurring collaborative tasks.

---

Conduct a structured cross-project code review between two Claude Code instances via the cross-review-mcp broker. Each agent reads its own codebase, reviews the peer's code, and engages in evidence-backed dialogue — with QSG scaling laws enforcing review quality through minimum bandwidth constraints and phase-gated progression.

---

---

---

Classify gate call variants in a minified JavaScript bundle. Covers context-window extraction around a flag occurrence, identification of 4–6 reader variants (sync boolean, sync config-object, bootstrap-aware TTL, truthy-only, async bootstrap, async bridge), default-value extraction (boolean / null / numeric / config-object literal), conjunction detection across `&&` predicates, kill-switch inversion detection, and production of a gate-mechanics record that feeds probe- feature-flag-state. Use when a flag's behavior cannot be inferred from its name alone, when the binary uses multiple reader libraries, or when config-object gates carry structured schemas distinct from boolean gates.

---

---

Establish Service Level Objectives (SLO), Service Level Indicators (SLI), and Service Level Agreements (SLA) with error budget tracking, burn rate alerts, and automated reporting using Prometheus and tools like Sloth or Pyrra. Use when defining reliability targets for customer-facing services, balancing feature velocity against system reliability through error budgets, migrating from arbitrary uptime goals to data-driven metrics, or implementing Site Reliability Engineering practices.

Deploy machine learning models to edge devices using Google AI Edge Gallery, TensorFlow Lite, ONNX Runtime, and MediaPipe. Covers model quantization (INT8/INT4), on-device inference with Gemma 4 models, Android/iOS deployment via AI Edge Gallery, hardware delegate selection (GPU/NPU/DSP), and performance benchmarking on constrained devices. Use when deploying models to mobile phones, IoT devices, or embedded systems where cloud inference is impractical due to latency, cost, or connectivity constraints.

Deploy machine learning models to production serving infrastructure using MLflow, BentoML, or Seldon Core with REST/gRPC endpoints, implement autoscaling, monitoring, and A/B testing capabilities for high-performance model inference at scale. Use when deploying trained models for real-time inference, setting up REST or gRPC prediction APIs, implementing autoscaling for variable load, running A/B tests between model versions, or migrating from batch to real-time inference.

---

---

---

Deploy applications to Kubernetes clusters using kubectl manifests for Deployments, Services, ConfigMaps, Secrets, and Ingress resources. Implement health checks, resource limits, rolling updates, and Helm chart packaging for production deployments. Use when deploying new applications to EKS, GKE, AKS, or self-hosted clusters, migrating from Docker Compose to container orchestration, implementing zero-downtime rolling updates, or setting up multi-environment deployments across dev, staging, and production.

---

---

---

---

Design terminal output for a CLI tool with chalk colors, Unicode glyphs, multiple verbosity levels (human, verbose, quiet, JSON), and consistent voice rules. Covers color palette selection, status indicator design, reporter function architecture, ceremony/narrative output variants, and cross-terminal compatibility. Use when building a new CLI reporter module, adding warm narrative output to an existing tool, standardizing output across multiple commands, or designing machine-readable JSON alongside human-readable text.

---

---

---

Design sustainable on-call rotations with balanced schedules, clear escalation policies, fatigue management, and handoff procedures. Minimize burnout while maintaining incident response coverage. Use when setting up on-call for the first time, scaling a team from 2-3 to 5+ engineers, addressing on-call burnout or alert fatigue, improving incident response times, or after a post-mortem identifies handoff issues.

---

---

---

Implement AI-powered anomaly detection for operational metrics using time series analysis (Isolation Forest, Prophet, LSTM), alert correlation, and root cause analysis. Reduce alert fatigue by intelligently identifying true anomalies in system metrics, logs, and traces. Use when operations teams are overwhelmed by alert volume, when detecting complex multi-metric anomalies beyond static thresholds, when seasonal patterns make thresholds ineffective, or when needing to predict issues proactively before they impact users.

---

---

---

---

---

---

Separate expensive observation from cheap decision-making in autonomous agent loops using a two-clock architecture. A fast clock accumulates data into a digest file; a slow clock reads the digest and acts only when something is pending. Idle cycles cost nothing because the action clock returns immediately after reading an empty digest. Use when building autonomous agents that must observe continuously but can only afford to act occasionally, when API or LLM costs dominate and most cycles have nothing to do, when designing cron-based agent architectures with observation and action phases, or when an existing heartbeat loop is too expensive because it calls the LLM on every tick.

Implement policy-as-code enforcement using OPA Gatekeeper or Kyverno to validate and mutate Kubernetes resources according to organizational policies. Covers constraint templates, admission control, audit mode, reporting violations, and integrating with CI/CD pipelines for shift-left policy validation. Use when enforcing resource configuration standards, preventing security misconfigurations such as privileged containers, ensuring compliance before deployment, standardizing naming conventions, or auditing existing cluster resources against policies.

Improve an existing R-based pictogram glyph for the visualization layer. Covers visual audit of the current glyph, diagnosis of specific issues (proportions, readability, glow balance), targeted modifications to the glyph function, re-rendering, and before/after comparison. Works for skill, agent, and team glyphs. Use when a glyph renders poorly at small sizes, its visual metaphor is unclear, it has proportion issues, the neon glow effect is unbalanced, or after adding new palettes or changing the rendering pipeline.

---

Assess an open-source agent framework for investment readiness by evaluating community health, supersession risk, architecture alignment, and governance sustainability. Produces a four-tier classification (INVEST / EVALUATE-FURTHER / CONTRIBUTE-CAUTIOUSLY / AVOID) to guide resource allocation decisions before committing engineering effort.

---

---

Evolve an existing agent definition by refining its persona in-place or creating an advanced variant. Covers assessing the current agent against best practices, gathering evolution requirements, choosing scope (refinement vs. variant), applying changes to skills, tools, capabilities, and limitations, updating version metadata, and synchronizing the registry and cross-references. Use when an agent's skills list is outdated, user feedback reveals capability gaps, tool requirements have changed, an advanced variant is needed alongside the original, or the agent's scope needs sharpening after real-world use.

Evolve SKILL.md files from agent execution traces using a three-stage pipeline: trajectory collection from observed runs, parallel multi-agent patch proposal for error and success analysis, and conflict-free consolidation of overlapping edits via prevalence-weighting. Based on the Trace2Skill methodology.

Evolve an existing skill by refining its content in-place or creating an advanced variant. Covers assessing the current skill, gathering evolution requirements, choosing scope (refinement vs. variant), applying changes, updating version metadata, and synchronizing the registry and cross-references. Use when a skill's procedure steps are outdated, user feedback reveals gaps, a skill needs a complexity upgrade, an advanced variant is needed alongside the original, or related skills are added and cross-references are stale.

Evolve an existing team composition by refining its structure in-place or creating a specialized variant. Covers assessing the current team against template and coordination patterns, gathering evolution requirements, choosing scope (adjust members, change coordination pattern, split/merge teams), applying changes to the team file and CONFIG block, updating version metadata, and synchronizing the registry and cross-references. Use when a team's member roster is outdated, coordination pattern no longer fits, user feedback reveals workflow gaps, a specialized variant is needed alongside the original, or agents have been added or removed from the library affecting team composition.

Structured procedure for expanding attention from focused single-domain mode to panoramic multi-domain awareness. The cognitive transition from "focused attention on one problem" to "unfocused attention encompassing all relevant domains simultaneously." Like Baars' Global Workspace — consciousness as broadcast rather than spotlight. Use after meditation has cleared noise, when a problem spans multiple domains that need to be perceived together, when single-domain analysis keeps missing cross-domain connections, or as the opening move before integrate-gestalt synthesizes what is perceived.

---

Communicate an integrated insight in a way that is accessible, actionable, and preserves the multi-domain nature of the understanding. Integrated insights are multi-dimensional — linearizing them risks losing the relationships that make them valuable. This skill provides a structured procedure for choosing the right form, expressing the gestalt with honest attribution, and inviting productive challenge. Use after integrate-gestalt has formed a cross-domain understanding that needs to be communicated to an audience — domain specialists, generalists, or decision-makers.

Apply the fail-early (fail-fast) pattern to detect and report errors at the earliest possible point. Covers input validation with guard clauses, meaningful error messages, assertion functions, and anti-patterns that silently swallow failures. Primary examples in R with general/polyglot guidance. Use when writing functions that accept external input, adding input validation before CRAN submission, refactoring code that silently produces wrong results, reviewing PRs for error-handling quality, or hardening internal APIs against invalid arguments.

Trademark filing procedures covering EUIPO (EU), USPTO (US), and WIPO Madrid Protocol (international). Walks through pre-filing conflict checks, Nice classification, descriptiveness assessment, mark type decisions, filing basis strategy, office-specific e-filing procedures, Madrid Protocol extension, post- filing monitoring, and open-source trademark policy drafting. Use after running screen-trademark to confirm the mark is clear, when ready to secure trademark rights in one or more jurisdictions.

---

---

---

---

---

Forecast infrastructure and application metrics using Prophet or statsmodels for capacity planning, cost optimization, and proactive scaling. Visualize predictions in Grafana and set up alerts for projected resource exhaustion. Use when forecasting infrastructure capacity needs for CPU, memory, or disk, planning hardware procurement for next quarter, predicting cost trends to optimize cloud spending, or setting up proactive scaling policies based on predicted load.

---

Format citations across academic styles (APA 7, Chicago, Vancouver, IEEE) using CSL processors and R tooling. Convert between citation styles, generate in-text citations and reference lists, and validate formatting against style guides using citeproc, knitcitations, and Quarto's built-in citation engine. Use when rendering a Quarto or R Markdown document with formatted citations, converting a bibliography between citation styles, generating a standalone reference list, or setting up citation infrastructure for a multi-document project.

---

---

---

---

---

---

---

---

---

---

---

---

Extract data from web pages using the scrapling Python library — select the appropriate fetcher tier (HTTP, stealth Chromium, or full browser automation) based on target site defenses, configure headless browsing, and extract structured data with CSS selectors. Use when WebFetch is insufficient for JS-rendered pages, anti-bot-protected sites, or structured multi-element extraction requiring DOM traversal.

---

---

---

---

---

---

---

---

---

Implement GitOps continuous delivery using Argo CD or Flux with app-of-apps pattern, automated sync policies, drift detection, and multi-environment promotion. Manage Kubernetes deployments declaratively from Git with automated reconciliation. Use when implementing declarative infrastructure management, migrating from imperative kubectl commands to Git-driven deployments, setting up multi-environment promotion workflows, enforcing code review gates for production, or meeting audit and compliance requirements.

---

Install skills, agents, and teams from agent-almanac into any supported agentic framework using the CLI. Covers framework detection, content search, installation with dependency resolution, health auditing, and manifest-based syncing. Use when setting up a new project with agentic capabilities, installing specific skills or entire domains, targeting multiple frameworks simultaneously, or maintaining a declarative manifest of installed content.

---

Instrument applications with OpenTelemetry for distributed tracing, including auto and manual instrumentation, context propagation, sampling strategies, and integration with Jaeger or Tempo. Use when debugging latency issues in distributed systems, understanding request flow across microservices, correlating traces with logs and metrics for root cause analysis, measuring end-to-end latency, or migrating from legacy tracing systems to OpenTelemetry.

Form a coherent gestalt — the whole that is more than the sum of its parts — from the panoramic perception produced by expand-awareness. Maps tensions and resonances between domains, identifies the emergent figure from the ground of multiple perspectives, tests the candidate whole for premature closure, and articulates the insight in a single sentence no single domain could have produced. Use after expand-awareness has surfaced raw multi-domain perception and before express-insight communicates the result.

---

---

---

---

---

---

---

---

Set up systematic data labeling workflows using Label Studio or similar tools. Implement quality controls, measure inter-annotator agreement, manage labeler teams, and integrate labeled data into ML training pipelines. Use when starting a supervised ML project that requires labeled training data, when model performance is limited by insufficient labeled examples, when labeling text, images, audio, or video, or when implementing active learning to prioritize the most valuable examples.

---

---

---

---

---

---

---

Create, merge, and deduplicate BibTeX bibliography files using R packages (RefManageR, bibtex). Parse .bib files into structured R objects, merge multiple bibliographies with deduplication by DOI or title similarity, generate entries from DOI/ISBN/arXiv ID, and export clean sorted .bib files. Use when creating a new .bib file for an R Markdown or Quarto project, merging bibliographies from multiple collaborators, deduplicating a .bib that has grown through copy-paste accumulation, or generating BibTeX entries programmatically from DOIs or other identifiers.

---

---

Manage an engagement buffer that ingests, prioritizes, rate-limits, deduplicates, and tracks state for incoming engagement items across platforms. Generates periodic digests and enforces cooldown periods. Composes with du-dum: du-dum sets the observation/action cadence, this skill manages the queue between beats.

---

Implement secure secrets management in Kubernetes using SealedSecrets for GitOps, External Secrets Operator for cloud secret managers, and rotation strategies. Handle TLS certificates, API keys, and credentials with encryption at rest and RBAC controls. Use when storing sensitive configuration for Kubernetes applications, implementing GitOps where secrets must be version-controlled, integrating with AWS Secrets Manager or Azure Key Vault, rotating credentials without downtime, or migrating from plaintext Secrets to encrypted solutions.

---

---

---

---

---

---

---

---

---

---

Establish and maintain longitudinal baselines of CLI binary contents across versions. Covers marker selection by category (API / identity / config / telemetry / flag / function), weighted scoring, threshold-based system-presence detection, and per-version baseline records. Use when tracking a feature's lifecycle across releases, when probing for dark-launched or removed capabilities, or when verifying that a scanning tool itself still catches known-good markers on old binaries.

---

Implement comprehensive model drift monitoring using Evidently AI, statistical tests (PSI, KS), and custom metrics to detect data drift and concept drift in production ML systems. Set up automated alerting and reporting workflows to catch degradation before it impacts business metrics. Use when production models show unexplained performance degradation, when new data distributions differ from training data, when seasonal shifts affect input features, or when regulatory requirements mandate model monitoring.

---

---

---

---

---

Implement cloud cost optimization strategies for Kubernetes workloads using tools like Kubecost for visibility, right-sizing recommendations, horizontal and vertical pod autoscaling, spot/preemptible instances, and resource quotas. Covers cost allocation, showback reporting, and continuous optimization practices. Use when cloud costs are growing without proportional business value, when resource requests are misaligned with actual usage, when manual scaling leads to over-provisioning, or when implementing showback and chargeback for internal cost accountability.

---

---

Orchestrate end-to-end machine learning pipelines using Prefect or Airflow with DAG construction, task dependencies, retry logic, scheduling, monitoring, and integration with MLflow, DVC, and feature stores for production ML workflows. Use when automating multi-step ML workflows from data ingestion to deployment, scheduling periodic model retraining, coordinating distributed training tasks, or managing retry logic and failure recovery across pipeline stages.

---

---

---

---

---

Apply software pipelining (double-buffering) to a tiled GPU kernel to overlap global memory loads with Tensor Core computation. Covers prologue/loop/epilogue restructuring, LDG-register vs cp.async (LDGSTS) variant selection based on compute/load ratio, shared memory budget verification against architecture-specific occupancy cliffs, and SASS-level verification of load/compute overlap.

Perform capacity planning using historical metrics and growth models. Use predict_linear for forecasting, identify resource constraints, calculate headroom, and recommend scaling actions before saturation. Use before seasonal traffic spikes or product launches, during quarterly capacity reviews, when resource utilization trends upward, or before budget planning cycles.

---

---

---

---

---

---

---

Contribute to OpenClaw ecosystem projects (OpenClaw, NemoClaw, NanoClaw) through a structured 9-step workflow: target verification, codebase exploration, parallel audit, finding cross-reference, and pull request creation. Emphasizes false positive prevention and project convention adherence.

---

---

---

---

---

---

Probe the runtime state of a named feature flag in a CLI binary. Covers the four-pronged evidence protocol (binary strings, live invocation, on-disk state, platform cache), the four-state classification (LIVE / DARK / INDETERMINATE / UNKNOWN), gate-vs-event disambiguation, conjunction-gate handling, and skill-substitution scenarios where a flag appears DARK but the capability is delivered by other means. Use when verifying whether a documented or inferred capability has rolled out, when auditing dark-launched features, or when a prior probe's conclusions need refreshing against a new binary version.

---

Provision and manage cloud infrastructure using Terraform with HCL modules, remote state backends, workspaces, and plan/apply workflow. Implement infrastructure as code patterns with variable management, output values, and state locking for team collaboration. Use when provisioning new cloud infrastructure, migrating from ClickOps or CloudFormation to declarative IaC, managing multi-environment infrastructure, versioning infrastructure changes alongside application code, or enforcing standards through reusable modules.

Audit, classify, and selectively forget stored memories. Covers memory enumeration and classification by type/age/access frequency, staleness detection for outdated references, fidelity checks using external anchors, a decision tree for selective deletion, counter-memory inoculation for failed strategies that would otherwise be re-derived, preemptive filtering rules for what should never become memories, and an audit trail so forgetting itself is reviewable. Use when memory has grown large and uncurated, when project state has shifted significantly since memories were written, when retrieval quality has degraded, or as periodic maintenance alongside manage-memory.

---

---

Read a CONTINUE_HERE.md continuation file at session start and resume from where the prior session left off. Covers detecting the file, assessing freshness, parsing the structured handoff, confirming the resumption plan with the user, and cleaning up after consumption. Optionally configures a SessionStart hook and CLAUDE.md instruction for automatic pickup. Use at the start of a session when a continuation file exists, when bootstrapping after an interrupted session, or when setting up automatic continuation detection.

---

---

Redact reverse-engineering findings for public disclosure while preserving methodology, generalizable patterns, and teaching value. Covers the private-vs-public repo split, deny-list pattern maintenance, orphan-commit publish pattern that prevents `git log` leaks, category-based redaction calibration (methodology/pattern/version-finding/internal), and the `check-redaction.sh`-style CI gate that blocks merges when a deny-listed pattern appears. Use when publishing findings about a CLI harness you don't own, when preparing upstream proposals to an unrelated project, or when archiving a private research repo for public reference.

---

---

Register trained models in MLflow Model Registry with version control, implement stage transitions (Staging, Production, Archived) with approval workflows, and manage model lineage with comprehensive metadata and deployment tracking. Use when promoting a trained model from experimentation to production, managing multiple model versions across development stages, implementing approval workflows for governance, rolling back to previous versions, or auditing model changes for compliance.

---

---

---

---

Run the viz pipeline to render icons from existing glyphs. Entry point for the viz subproject covering palette generation, data building, manifest creation, and icon rendering for skills, agents, and teams. Always use build.sh as the pipeline entry point — never call Rscript directly.

---

---

---

---

---

---

---

---

---

---

---

---

---

---

---

Escalate blocked scraping campaigns with provider-neutral proxy rotation — decide between datacenter, residential, and mobile pools, integrate rotation with scrapling, configure session stickiness for stateful flows, monitor cost and health, and stay inside legal and ethical boundaries. Use as the next step after `headless-web-scraping` client-side stealth (StealthyFetcher, rate limiting, robots.txt) is insufficient and traffic is legitimate.

Design and execute A/B tests for ML models in production using traffic splitting, statistical significance testing, and canary/shadow deployment strategies. Measure performance differences and make data-driven decisions about model rollout. Use when validating a new model version before full rollout, comparing candidate models trained with different algorithms, measuring business metric impact of model changes, or when regulatory requirements mandate gradual rollout.

Design and execute chaos engineering experiments using Litmus or Chaos Mesh. Test system resilience through controlled fault injection, validate hypothesis-driven tests, and improve failure recovery. Use before major product launches, after architecture changes to validate resilience, during GameDays or disaster recovery drills, to validate assumptions about failure modes, or as part of an SRE maturity program.

---

Scaffold a new CLI command using Commander.js with options, action handler, three output modes (human-readable, quiet, JSON), and optional ceremony variant. Covers command naming, option design, shared context patterns, error handling, and integration testing. Use when adding a command to an existing Commander.js CLI, designing a new CLI tool from scratch, or standardizing command structure across a multi-command CLI.

---

---

---

---

Screen a proposed trademark for conflicts and distinctiveness before filing. Covers trademark database searches (TMview, WIPO Global Brand Database, USPTO TESS), distinctiveness analysis using the Abercrombie spectrum, likelihood of confusion assessment using DuPont factors and EUIPO relative grounds, common law rights evaluation, and goods/services overlap analysis. Produces a conflict report with a risk matrix. Use before adopting a new brand name, logo, or slogan — distinct from patent prior art search, which uses different databases, legal frameworks, and analysis methods.

---

---

---

---

---

Configure automated machine learning pipelines using Optuna or Ray Tune for hyperparameter optimization. Implement efficient search strategies (Hyperband, ASHA), define search spaces, and set up early stopping to find optimal model configurations with minimal manual tuning. Use when starting a new ML project and needing to quickly find good configurations, retraining with new data and re-optimizing hyperparameters, comparing multiple algorithms, or when the team lacks deep expertise in specific algorithm hyperparameters.

---

Configure container image registries including GitHub Container Registry (ghcr.io), Docker Hub, and Harbor with automated image scanning, tagging strategies, retention policies, and CI/CD integration for secure image distribution. Use when setting up a private container registry, migrating from Docker Hub to self-hosted registries, implementing vulnerability scanning in CI/CD pipelines, managing multi-architecture images, enforcing image signing, or configuring automatic cleanup and retention policies.

---

---

---

Set up a local Kubernetes development environment using kind, k3d, or minikube for fast inner-loop development. Covers cluster creation, ingress configuration, local registry setup, and integration with development tools like Skaffold and Tilt for automatic rebuild and redeploy workflows. Use when needing a local Kubernetes environment for development, testing manifests or Helm charts before production deployment, wanting fast automatic rebuild-and-redeploy cycles, or learning Kubernetes without cloud costs.

Configure Prometheus for time-series metrics collection, including scrape configurations, service discovery, recording rules, and federation patterns for multi-cluster deployments. Use when setting up centralized metrics collection for microservices, implementing time-series monitoring for application and infrastructure, establishing a foundation for SLO/SLI tracking and alerting, or migrating from legacy monitoring solutions to a modern observability stack.

---

Deploy and configure a service mesh (Istio or Linkerd) to enable secure service-to-service communication, traffic management, observability, and policy enforcement in Kubernetes clusters. Covers installation, mTLS configuration, traffic routing, circuit breaking, and integration with monitoring tools. Use when microservices need encrypted service-to-service communication, fine-grained traffic control for canary or A/B deployments, observability across all service interactions without application changes, or consistent circuit breaking and retry policies.

---

Configure external uptime monitoring using Blackbox Exporter and Prometheus. Implement SSL certificate monitoring, HTTP endpoint health checks, and status pages for customer-facing visibility. Use when monitoring customer-facing endpoints such as APIs and websites, tracking SSL certificate expiration, validating service availability from multiple regions, creating public status pages, or meeting SLA requirements for uptime reporting.

---

---

---

---

---

---

---

---

---

---

---

---

---

---

Bulk-extract every candidate flag from a binary namespace, build an extraction inventory with occurrence counts and call-type tags, cross- reference against a documented set, and track completeness across probe campaigns until the undocumented remainder reaches zero. Covers namespace prefix harvesting, gate-vs-telemetry disambiguation at the call-site level, completeness metrics, DEFAULT-TRUE population reporting, and a final completion confirmation scan. Use upstream of probe-feature-flag- state when you need a complete catalog rather than a sample, or when a prior wave-based campaign needs a verifiable end condition.

---

---

---

---

Write integration tests for a Node.js CLI application using the built-in node:test module. Covers the exec helper pattern, output assertions, filesystem state verification, cleanup hooks, JSON output parsing, error case testing, and state restoration after destructive tests. Use when adding tests to an existing CLI, testing a new command, verifying adapter behavior across frameworks, or setting up CI for a CLI tool.

---

---

---

Set up MLflow tracking server for experiment management, configure autologging for popular ML frameworks, compare runs with metrics and visualizations, and manage artifacts in remote storage backends for reproducible machine learning workflows. Use when starting a new ML project that requires experiment tracking, migrating from manual logs to automated tracking, comparing multiple training runs systematically, or building reproducible ML workflows with full lineage tracking.

---

---

---

---

---

Launch all available agents in parallel waves for open-ended hypothesis generation on problems where the correct domain is unknown. Use when facing a cross-domain problem with no clear starting point, when single-agent approaches have stalled, or when diverse perspectives are more valuable than deep expertise. Produces a ranked hypothesis set with convergence analysis and adversarial refinement.

---

---

---

---

Check BibTeX entries for completeness, DOI resolution, and broken links. Verify required fields per entry type (article, book, inproceedings), resolve and validate DOIs via the CrossRef API, check URL accessibility, and flag duplicate entries, missing abstracts, and inconsistent formatting. Use when preparing a manuscript bibliography for journal submission, auditing a shared .bib file before a project milestone, after merging bibliographies from multiple sources, when citations render incorrectly, or as a CI check on version-controlled .bib files.

---

---

Version machine learning datasets using DVC (Data Version Control) with remote storage backends, build reproducible data pipelines with dependency tracking, integrate with Git workflows, and ensure data lineage for model reproducibility. Use when versioning large datasets that do not fit in Git, tracking data changes alongside code changes, ensuring ML experiment reproducibility, sharing datasets across team members, or auditing data lineage for compliance requirements.

---

---

Write a CONTINUE_HERE.md file capturing current session state so a fresh Claude Code session can pick up where this one left off. Covers assessing recent work, structuring the continuation file with objective, completed, in-progress, next-steps, and context sections, and verifying the file is actionable. Use when ending a session with unfinished work, handing off context between sessions, or preserving task state that git alone cannot capture.

Create production-ready Helm charts for Kubernetes application deployment with templating, values management, chart dependencies, hooks, and testing. Covers chart structure, Go template syntax, values.yaml design, chart repositories, versioning, and best practices for maintainable and reusable charts. Use when packaging a Kubernetes application for repeatable deployments, parameterizing manifests for multiple environments, managing complex multi-component applications with dependencies, or standardizing deployment practices with versioned rollback capability across teams.

Create structured incident runbooks with diagnostic steps, resolution procedures, escalation paths, and communication templates for effective incident response. Use when documenting response procedures for recurring alerts, standardizing incident response across an on-call rotation, reducing MTTR with clear diagnostic steps, creating training materials for new team members, or linking alert annotations directly to resolution procedures.

---

---

---

---

---