deniz-dotnet-aspire
claudecodexDeniz's curated .NET Aspire skills
10 skills
Claude Code and Codex plugins their authors filed under "productivity". All categories.
Deniz's curated .NET Aspire skills
10 skills
Ready-made Claude Skills for partnership and go-to-market teams. Brings partner overlap, co-sell context, and ecosystem signals from Crossbeam into revenue workflows.
10 skills
即梦AI创作工作流插件 - 支持视频/图片提示词生成等功能
2 skills · 6 commands · 2 agents
Capture predictions, choices and commitments with a confidence and a date; an adversary argues the other side at full tier; scripts score the record (Brier, calibration, release rate, process, drift); the weekly review forces a choice on every open item. Signals about the record, never verdicts about the person. State lives in ~/.cairn or a project's docs/cairn/.
6 skills · 3 agents · hooks
Keeps large tool output out of Claude's context window. Vendored copy of mksglu/context-mode v1.0.169 (Elastic License 2.0).
8 skills · 1 MCP server · hooks
A structured personal knowledge vault. Human-accessible via Obsidian; AI-operated through the ContextOS MCP server from Claude Code and Codex.
10 skills
Data platform copilot — connect your data stack and get analyst, engineer, and devops agents that understand your environment
10 skills
Give any AI agent a real, routable, verifiable network identity - a Whisper IPv6 /128 it egresses from, and keyless tools to verify any peer.
1 skill · 7 commands · 1 MCP server · hooks
Multi-agent bug hunting with 8 parallel specialist agents, deduplication, regression guard, and prioritized reporting
1 command · 9 agents
Conversations into plans, tickets, docs, and repeatable setup.
10 skills
Editorial design, prototypes, decks, documents, dashboards, and critique
9 skills · 1 command
The Hoplite authoring skills — glossary, spec, decision, notes, journal, todo, triage, domain modeling — plus the frontmatter-validation hook and the locked specs they enforce.
9 skills · hooks
Agent 讲解·演示·协作能力包:概念讲解编排、D2 结构图、Excalidraw 白板、Mermaid 图、隔离任务上下文协议与三大工作流命令;随插件自动注册 p2m-server MCP 门面(render / validate / preview / viewer / status)。
5 skills · 4 commands · 1 MCP server
Ratchet turns coding sessions into self-evolving wisdom for agents and developers. It automatically generates Skills and Strategies from past sessions, curates the right workflows and Cheatmaps from the Wisdom DB based on user intent, and enhances existing Skills through evaluation and ROI analysis. BYOK model with local extraction pipeline.
9 skills · hooks
Domain research skills.
10 skills
Markdown note management with Notebook→Section→Page hierarchy and unified slash commands. Import from OneNote, auto-learn templates from your notes, archive old content, and generate an AGENTS.md workspace manifest. Works with any agent — Copilot, Claude Code, or Codex.
1 skill · 9 commands
Workflows whose subject is AI agents: handoff debriefs, Project seeding, space consolidation, AI entity design, research source rules, routing tasks to other AI assistants, driving chat assistants in the browser, YouTube ingestion via Gemini, scouting existing skills.
10 skills
Day-planning skills: the next task picked and started for one client, tomorrow's work pulled forward when today is clear, day and week triage that proposes rather than writes, an unattended morning plan that fills the week against each client's hours and only moves items, and the day's plate across all clients from the Obsidian dashboard, with weekend fall-forward and tracker-first sourcing
10 skills
Machine and infrastructure administration for agent fleets: readiness, inventory, parity, packages, dotfiles, auth, SSH transport, privileged installs, and UniFi.
10 skills
Four procedures to find, save, review and set up shared project knowledge, with one reference manual, three short startup reads, topic records and interrupted-save recovery.
10 skills
Dev workflow & meta toolkit - PAUL loop, seed incubator, brainstormers, research, arXiv prior-art (neuroarxiv), project-approach advisory, CLI automation, gstack, CLAUDE.md config, ecosystem directory
10 skills
Interviews you into a use-case document that an independent agent then reviews adversarially, breaks the feature down into commit-sized implementation plans, and executes them with subagents sized to each plan's complexity. Run /impl-flow:spec (design+plan) and /impl-flow:implement independently, or /impl-flow:all for the full pipeline in one go. For changes to something that already exists, /impl-flow:fix is a lighter fix mode that revises the existing use-case document and hunts for what a partial fix would leave behind.
7 commands · 3 agents
Plain-language skills for coding agents, for the person running development who does not read code: session reports, decisions, next steps and the whole picture.
8 skills · 1 command · hooks
Self-contained workflow for administering and troubleshooting remote systems (Linux/SSH, cloud, containers/k8s, appliances): loads FLEET.md plus the INCIDENTS / CHANGELOG / RUNBOOKS journals at session start, re-injects ops standing rules per turn, flags destructive Bash commands with an out-of-band audit trail, gates session exit on a journaling decision, and ships the fleet-init / fleet-status / troubleshooting-loop / fleet-onboard / incident-record / runbook-capture skills.
9 skills · hooks
開発支援スキル (コードコメント追加、行動計画作成、並列タスク実行、HTML ドキュメント生成、フィードバック検証・反映、ローカルレビュー、orchestrator-workers 実行、advisor 相談)
8 skills · 2 agents
Manual-only decision procedures. Each skill runs one published protocol against a judgment already on the table, structuring it without making it. Ships /reversible (one-way and two-way doors), /options (WRAP widening), /tradeoff (Kepner-Tregoe and Pugh), /hypotheses (ACH), /premortem (Klein), /redteam (adversarial review), /fermi (order-of-magnitude estimation), /base-rate (reference class forecasting), /quit (kill criteria), and /adr (Nygard decision records).
10 skills
Accurate knowledge of OneLiteFeather's internal Minestom libraries and tooling (Cyano testing, Gradle conventions, BOM hierarchy, Guira setup flows, Aves utilities, Xerus minigame core, Pica dialogs, Coris floors/rooms/doors), too internal/new to be in general training data.
10 skills
Scheduled personal briefings rendered as self-contained HTML on the owner's own machine. /commitment-calendar builds a rolling N-day commitment grid from the calendar, the inbox, and every configured work surface — team board, repositories, deployments and domain renewals, scheduled social posts, content in flight, outreach queues, search and competitor watch, product experiments, invoices, payroll, contracts, pipeline commitments, decision records awaiting sign-off, and the scheduled tasks themselves — flagging real clashes, duplicate invites, location clashes and deadlines with no prep blocked. /morning-brief renders today's version. /crm-sync-health checks the CRM sync pipeline, sweeps orphaned runs and releases a dead-lettered source once a day, and costs one line on a healthy day. /briefings-setup writes the owner config and registers the recurring tasks. Read-only across every other surface; a source that could not be read is reported as not checked, never as empty.
5 skills · 5 commands
Cowork project lifecycle for MoxyWolf projects. /init-project scaffolds tailored Project Instructions. /session-start auto-resolves the project — explicit argument, then launch directory, then the most recently active project, never a picker — mounts the three standard roots, reads the previous session's handoff (cross-checked against current team-shared memory so stale procedural reminders get flagged STALE rather than silently propagated), reads GitHub/Projects/_shared-memory/INDEX.md and surfaces it as a 'Team-shared rules in effect' briefing section, and probes the team-shared OpenRouter key (DR-010). /session-end asks who authored the handoff and whether it's a baton-pass, writes the canonical handoff with `author` / `for` frontmatter, requires every general behavioral reminder in the procedural-reminders section to cite a currently-existing team-shared memory file (or be tagged session-specific), supports --author / --handoff-to / --solo, refreshes each writable repo's README.md, then runs /obsidian-update. /init-project saves the full instructions on disk as the single source of truth and pastes only a thin loader stub (mounts + file-write-path override + read-the-file-first imperative) into Cowork settings, so the embedded copy never drifts out of sync with the on-disk file (DR-010). The stub skeleton and on-disk reconciliation directives live in a vault spec that both /init-project and /refresh-project-instructions read at run time, so changing a project-instruction rule is a vault edit plus a per-project /refresh-project-instructions run, never a plugin update. v0.22.0: folder access is fully mode-agnostic across online (cloud) and offline (on-computer) — each capability is detected by tool presence with a filesystem-probe fallback (detect via get_device_info or by listing the path; mount via request_cowork_directory when present, else prompt Add folder; pick via native picker when present, else device_list_dir + AskUserQuestion chips), so nothing hard-dep
3 skills · 6 commands · hooks
Manage Spotify ad campaigns with natural language. Create campaigns, ad sets, ads, pull reports, and handle OAuth — all through conversation.
8 skills · 1 agent · hooks