ldaume
- 30 skills
- 0 followers
- 9 hours ago last updated
- ▌ Pnpm · ldaumeHandles pnpm package management, workspaces, lockfiles, scripts, catalogs, overrides, patches, Corepack, and dependency troubleshooting. Use when running pnpm commands, configuring pnpm workspaces, fixing strict dependency resolution, managing packageManager versions, or updating dependencies in Node.js projects.
- ▌ API Design · ldaumeDesigns and reviews REST, GraphQL, and internal APIs with explicit contracts, validation, errors, pagination, versioning, documentation, auth, and compatibility. Use when creating endpoints, changing API contracts, writing OpenAPI docs, reviewing GraphQL schemas, or refactoring service boundaries.
- ▌ Meilisearch · ldaumeDesigns, configures, and debugs Meilisearch indexing, search settings, filters, ranking rules, synonyms, typo tolerance, multi-tenant indexes, and reindex/backfill flows. Use when adding Meilisearch, changing search relevance, debugging missing results, tuning facets, or operating search indexes.
- ▌ Backend Craft · ldaumeGuides backend architecture, TypeScript boundaries, APIs, data models, PocketBase, Flue workflows, BullMQ queues, workers, AI integrations, provider secrets, security, tests, and operability. Use when designing or changing backend systems, runtime validation, migrations, job queues, workflow orchestration, LLM outputs, or deployment-sensitive code.
- ▌ Product Craft · ldaume bundleShapes evidence-driven product investments, value-defined issues, honest Now/Next/Later/Never roadmaps, discovery, UX, architecture, delivery, and learning. Use when turning signals into bets, deciding whether or when work deserves investment, planning product work, or coordinating product and engineering trade-offs without backlog or milestone theater.
- ▌ Agent Sync · ldaumeProactively evolves repository harness artifacts, review loops, currentness, and agent routing across sessions and tools from durable evidence. Use when starting or completing a session, after toolchain, model, pricing, or domain changes, when repeated friction reveals missing context or checks, or when instructions, orchestration, Skills, Rules, Hooks, MCP, or learnings may be stale.
- ▌ Start Gate · ldaumePre-work gate for code changes - inspect the working tree, preserve foreign work in progress, get the default branch current, and claim an isolated branch or worktree before the first edit. Use when starting a task, resuming a session, picking up an issue, or when the state of the checkout is unknown.
- ▌ Frontend Craft · ldaumeGuides frontend, UX, UI, responsive design, microcopy, accessibility, feedback states, and AI-assisted product surfaces with production-grade taste. Use when designing, building, reviewing, or polishing screens, React islands/components, Tailwind UI, mobile flows, loading/empty/error states, or user-facing copy.
- ▌ Gitea Tea · ldaumeUses the tea CLI to inspect and operate Gitea repositories, issues, pull requests, releases, branches, labels, and workflow-adjacent metadata from the terminal. Use when the user mentions tea, gitea cli, terminal-based Gitea operations, PR inspection, issue triage, release checks, or repository automation against Gitea.
- ▌ Write A Skill · ldaumeCreates, refactors, packages, and verifies portable Agent Skills. Use when changing SKILL.md, skill descriptions, bundled references or scripts, repository indexes, install behavior, or activation quality across Cursor, Codex, Claude Code, Gemini, and the skills CLI.
- ▌ Update Harness · ldaume bundleChecks, installs, updates, and cleans repository or cross-repository harness Skills across project, private organization or team, public, user, and global scopes. Use for Skill installation, updates, synchronization, duplicate or conflicting Skill cleanup, missing public complements, Renovate Skill changes, source federation, or stale harness dependencies and platform assumptions.
- ▌ Testing Strategies · ldaumePlans pragmatic test coverage across unit, integration, contract, e2e, visual, accessibility, performance, and smoke tests based on risk and behavior. Use when deciding what to test, designing a test pyramid, adding coverage for a feature, reducing flaky tests, or reviewing QA strategy.
- ▌ Completion Gate · ldaumePre-finish review gate for code changes - correctness, patterns, security, tests, verification, and checking the deployment a change targets before claiming work complete. Use when finishing a task, creating commits, opening PRs, merging, or preparing to claim work complete.
- ▌ Scaffold Harness · ldaume bundleAudits and upgrades repositories to a reliable, maturity-appropriate Agentic Engineering harness without overwriting local truth. Use when bootstrapping or repairing agent instructions, designing local/MCP/RAG context routing, controlling token bloat, defining review, model, or quality gates, coordinating repositories, or evaluating agents for CI, overnight, specialist, chat, and observability work.
- ▌ Gitea Actions · ldaumeDesigns, debugs, secures, and reviews Gitea Actions workflows, act_runner setups, runner labels, tokens, secrets, variables, cache, artifacts, and GitHub Actions compatibility. Use when working with .gitea/workflows, Gitea CI, act_runner, GITEA_TOKEN permissions, self-hosted runners, workflow migration, sparse-checkout CI helpers, docker/build-push-action on Gitea, step summaries, or Gitea-specific CI failures.
- ▌ Coding Discipline · ldaumeEnforces clarity, surgical diffs, simplicity, and goal-driven TDD for any codebase. Use when implementing features, fixing bugs, refactoring, or making any code change.
- ▌ Scaffold Monorepo · ldaume bundleScaffolds a pnpm TypeScript monorepo with verify gates, knip, pre-commit hooks, pinned CI, and Renovate using current toolchain versions at scaffold time. Use when creating a new monorepo, bootstrapping apps/packages layout, or aligning a greenfield repo with progressive-pragmatic craftsmanship golden paths for GitHub Actions or Gitea Actions.
- ▌ Run Product Engineering · ldaume bundleRuns an evidence-driven product engineering system from signals and ideas through triage, problem framing, discovery, bets, small-batch implementation, validation, release, observability, incidents, bug fixing, outcome review, and evolution. Use when coordinating an end-to-end product lifecycle, designing a product operating model, connecting discovery with delivery and operations, or delegating bounded L5-L7 value streams to agents.
- ▌ Ansible Automation · ldaumeDesigns, reviews, and troubleshoots Ansible inventories, playbooks, roles, variables, vault usage, idempotency, and server automation. Use when managing infrastructure with Ansible, writing playbooks, organizing roles, handling SSH inventories, templating configs, or making repeatable operations safe.
- ▌ Playwright Best Practices · ldaumeWrites, reviews, and debugs Playwright tests for web apps with stable locators, fixtures, auth setup, API mocking, trace debugging, responsive checks, accessibility assertions, and CI reliability. Use when creating Playwright tests, fixing flaky e2e tests, debugging traces, testing UI flows, or configuring browser tests.
- ▌ Documentation And Adrs · ldaumeWrites and updates durable project documentation, decision records, runbooks, and architecture notes for a clear reader or operational purpose. Use when making architecture decisions, changing public APIs, adding operational procedures, documenting trade-offs, or deciding whether a change needs an ADR.
- ▌ Build Autonomous Agents · ldaume bundleDesigns and implements bounded autonomous agent systems for product features and software or product SDLC automation with runtime-neutral contracts. Use when building an agent, finite workflow, subagent, scheduled or overnight job, CI agent, chat or channel integration, observable agent service, adding Flue to an application, or migrating an older Flue integration.
- ▌ Grill Harness With Docs · ldaumeGrounds agent work in authoritative evidence, establishes shared understanding, critiques resolved material decisions and changes with a fresh agent, and resolves genuinely open decisions with a human. Use for resolved material work needing independent critique or for any product, domain, engineering, architecture, operations, security, compliance, or harness topic with unresolved intent, semantics, authority, consequential trade-offs, or material risk.
- ▌ Integrate Product Compliance · ldaume bundleIntegrates confirmed security, trust, and compliance scope into product engineering through risk, controls and policy as code, evidence, release policy, and effectiveness review. Use when work touches ISO/IEC 27001, TISAX or VDA ISA, PCI DSS, customer security commitments, regulated or sensitive data, payment flows, compliance as code, policy as code, machine-readable controls, audit evidence, control exceptions, or risk-based delivery.
- ▌ Learn Agentic Engineering · ldaume bundleActs as a self-directed Agentic Engineering mentor for software developers from first agent-assisted tasks through L7 product-system autonomy. Use when learning a concept, asking questions, practicing in a real repository, assessing maturity, planning a learning path, preparing coaching, diagnosing confusion or blockers, or seeking cause-and-effect guidance about context, DDD, TDD, Rules, AGENTS.md, Skills, Hooks, MCP, memory, orchestration, evals, governance, and human oversight.
- ▌ Multi Stage Dockerfile · ldaumeCreates and reviews secure, cache-friendly multi-stage Dockerfiles for Node, web, backend, and static-site projects. Use when writing Dockerfiles, optimizing image size, separating build/runtime stages, hardening containers, fixing slow Docker builds, or preparing production images.
- ▌ Deliver Dependency Upgrades · ldaumeAssess and deliver dependency version changes from a bot signal through source-backed impact analysis, adaptation, protected integration, rollout, observation, rollback, and linked feature-opportunity capture. Use when a dependency bump must pass through its bounded production lifecycle. Do not use for updater configuration, managed Skill pins, isolated CI repair, or release-note summaries.
- ▌ Secure Linux Web Hosting · ldaumeGuides secure setup and review of Linux web servers for self-hosted apps, static sites, reverse proxies, SSH, firewalls, TLS, backups, updates, logs, and basic hardening. Use when provisioning a VPS, hardening a server, configuring Nginx/Caddy/Traefik, deploying web apps, or reviewing self-hosting security.
- ▌ Scaffold Distributed Context · ldaume bundleEstablishes the smallest reliable domain-context system across repositories without confusing repositories, bounded contexts, generated graphs, or agent instructions with canonical truth. Use when agents must reason across services, applications, shells, microfrontends, or code hosts; when creating a cross-repository context map, domain glossary, contracts, or projections; or when evaluating Graphify, Zoekt, Sourcegraph, Backstage, or policy tooling for context discovery and governance.
- ▌ Manage Infrastructure As Code · ldaumeManages infrastructure through reviewable desired state, GitOps or GitOps-near reconciliation, plans, protected state, policy checks, controlled apply, drift detection, and recovery. Use for infrastructure as code, IaC, Terraform, OpenTofu, Pulumi, CloudFormation, Bicep, declarative infrastructure, GitOps, GitOps-near workflows, plan/apply, state backends, existing-resource imports, infrastructure drift, or infrastructure policy gates.