/products/skills · 31 skills

Skills your agent loads when the job calls for one

Each skill is a small, documented playbook — a specific job with a clear trigger. Every description below comes straight from the canonical SKILL.md in its plugin repo; this page just formats them.

Install via marketplace Browse solutions

/plugins

Skills by plugin

Install a provider first, then let its focused skills load when the job calls for them.

obsidian-agent30 skills

obsidian-agent

Claude Code skills for serious work inside an Obsidian vault. Pair it with companion-for-claude for the in-vault interface.

Companion for Claude Source

mlx-agent1 skill

mlx-agent

Local-model discovery and configuration for Apple Silicon through MLX Agent.

MLX Agent Source

/knowledge

Knowledge & synthesis

Answer questions from the vault, connect ideas, and digest sources — grounded and cited.

vault-synthesisobsidian-agent: Use when answering "what do I know about X", synthesizing across many notes, producing a grounded topic summary, or building a literature/research digest from an Obsidian vault.
connection-finderobsidian-agent: Use when looking for non-obvious connections between notes, surfacing related-but-unlinked notes, or finding serendipitous links across an Obsidian vault.
source-digestobsidian-agent: Use when digesting research sources or papers, extracting claims from notes tagged as sources, or building an evidence or comparison table from reference notes.
research-workbenchobsidian-agent: Use when framing a serious research project, capturing sources and evidence, building supported claims, auditing provenance, or producing an evidence-backed outline.

/hygiene

Vault hygiene

Keep tags, links, structure, and metadata consistent as the vault grows.

consistent-taggingobsidian-agent: Use when tagging notes, applying tags to new or untagged notes, or cleaning up tag sprawl and inconsistency in an Obsidian vault.
wikilink-weaverobsidian-agent: Use when connecting notes, finding missing links between notes, surfacing orphan notes, or strengthening the link graph in an Obsidian vault.
moc-builderobsidian-agent: Use when building or refreshing a Map of Content (MOC), creating an index or hub note for a topic or folder, or organizing related notes under one navigational note.
frontmatter-normalizerobsidian-agent: Use when enforcing a consistent frontmatter schema across notes, auditing or normalizing note metadata, or fixing inconsistent frontmatter fields in a folder.
note-splitterobsidian-agent: Use when a note covers too many topics and should be split into atomic notes, breaking up a bloated note, or extracting sections into their own linked notes.
dedup-mergeobsidian-agent: Use when finding and merging duplicate or near-duplicate notes, consolidating notes that cover the same thing, or cleaning up redundant notes in an Obsidian vault.

/writing

Writing

Turn outlines, captures, and recent activity into finished, linked notes.

outline-to-draftobsidian-agent: Use when expanding an outline or stub note into a full draft, fleshing bullet points into prose, or drafting longer-form writing grounded in vault context and the user's voice.
daily-rollupobsidian-agent: Use when summarizing recent or daily notes, creating a weekly or periodic review, or pulling decisions and open tasks out of recent vault activity.
session-to-noteobsidian-agent: Use when distilling or saving a Claude session into the Obsidian vault, capturing what was decided or learned in a session, or turning session memory into a persistent, linked knowledge note.
meeting-cleanupobsidian-agent: Use when turning raw meeting notes, voice memos, or messy capture into a structured note with decisions, action items, and attendees.
summarize-and-linkobsidian-agent: Use when adding a TL;DR or summary to a long note, condensing a note, or surfacing and linking the key concepts inside a note.

/build

Build, spec & handoff

Move from plan to spec to build, track progress honestly, and hand off to cloud sessions.

plan-to-specobsidian-agent: Use when turning a planning note into a build spec, preparing a note for handoff to a Claude Code build, or creating a spec and tracker from a plan.
tracker-driverobsidian-agent: Use when driving a spec-based build against the vault and updating its tracker note, or reporting build progress honestly to a tracker as tasks complete.
build-retrospectiveobsidian-agent: Use when a spec-based build or tracker is complete and a retrospective is needed, or when closing out a build against its spec.
task-harvesterobsidian-agent: Use when collecting open tasks or todos scattered across notes, building a consolidated action list, or surfacing all unchecked checkbox or task-tagged items in an Obsidian vault.
cloud-replyobsidian-agent: Use when running as a dispatched cloud Claude Code session against an Obsidian vault repo — do the task, write the result back as a reply note, and open a PR so Companion can pull it into the vault on any device.

/advisor

Advisor personas

manifest-* roadmaps: survey the vault through a persona and produce a prioritized plan.

manifest-vaultobsidian-agent: Use when asked to optimize, clean up, audit, or improve an Obsidian vault, assess vault health, or find structural problems in a vault.
manifest-pmobsidian-agent: Use when prioritizing project work from vault notes, deciding what to build next, planning client-facing deliverables, or producing a roadmap or status from project notes.
manifest-infraobsidian-agent: Use when proposing infrastructure or system designs from architecture notes, reviewing a system's design, or planning infra and scaling moves from vault notes.
manifest-featureobsidian-agent: Use when proposing product features from idea, feedback, or spec notes, deciding what to build from user feedback, or prioritizing a feature backlog from vault notes.
manifest-contentobsidian-agent: Use when turning vault knowledge into publishable content, planning a content calendar from your notes, or deciding what to write as a blog post, doc, or thread.
manifest-riskobsidian-agent: Use when surfacing risks, blockers, contradictions, or single points of failure across project notes, or building a risk register from the vault.
manifest-researchobsidian-agent: Use when identifying knowledge gaps, deciding what to research next, or planning a research agenda from what an Obsidian vault already covers.

/foundations

Foundations

The shared discipline every skill builds on — grounding, routines, and the artifact system.

vault-groundingobsidian-agent: Use when reading, citing, writing, or editing notes in an Obsidian vault over the MCP bridge — any time output references vault content, links notes, or modifies the vault.
vault-routinesobsidian-agent: Use when a vault task produces recurring value, when the user asks to automate or repeat a vault workflow on a schedule, or after delivering a result the user will want regularly.
note-to-artifactobsidian-agent: Use when turning notes, data, or a plan into a beautiful, self-contained HTML artifact (report, dashboard, plan, table, or diagram) in the Anthropic-flavored "HTML is all you need" style — especially for Obsidian notes rendered in a ```claude-html``` block.

/local-models

Local models

Discover, verify, and wire local MLX-optimized models on Apple Silicon into a coding agent.

mlx-scoutmlx-agent: Discover MLX-optimized models on HuggingFace suited to this Apple Silicon host, routed by role (general, coding, reasoning, vision, embedding).