Agent skills
Skills for AI coding agents, each with what it costs to load, how much of it is scripts and what it touches on your machine.
How to read a skill
What it costs to keep
Idle in context, in use when it runs, and up to with its linked files.
Scripts or instructions
Script-led, mixed or model-led: how much the model has to work out itself.
What it can reach
Shell, network, file writes, git and credentials, declared or detected.
Quality, security, repo health
Out of 100, each with what it is made of. Turn a card over to see them.
Sepia Refactor
3,912A fixed entry to the Sepia writing skill that runs only the refactor operation, listing defects first and then fixing them in place with minimal edits.
Sepia Review
3,912A fixed entry to the Sepia writing skill that runs only the review operation, reporting AI tells with quoted evidence and leaving your text unchanged.
Sepia Write
3,912A fixed entry to the Sepia writing skill that runs only the write operation, drafting new fiction or professional prose after reading the domain rules.
Agentic Video Understanding
3,862Give it a long video or audio file and a one-sentence goal. It returns timestamped moments with quotes, a reason and a confidence value, and does nothing else.
Closed-Loop Analytics Upgrade
3,862Check whether a marketing change worked by pulling analytics after the change window, comparing baseline to candidate, and patching the playbook from the evidence.
Content Eval
3,862Turn podcast transcripts, meeting notes, sales calls and competitor scans into 20 to 30 content ideas, score them with a seven-expert panel and get a ranked menu.
Growth Signal Charts
3,862Turns your original content and measured audience signals into sourced, chart-only social posts and carousels in a fixed Desk Edition style.
Monday Reel Ideas
3,862A weekly run that finds a creator's 3x outlier posts and verified outside breakouts, then writes three scripted short-form video ideas in one document.
Personal Strategic Signal Intelligence
3,862Reads your bookmarks, notes and applied work, separates attention from conviction, and produces source-cited decision briefs, contradictions and experiments.
Reimagine Work
3,862A weekly review of an agent fleet that finds where work waits on a human, ranks fixes and sets up a shadow owner judge, while every hard gate stays human.
shortform-format-library
3,862A workflow for keeping a library of short-form video formats and matching each content idea to a format, a recording outline and an edit brief.
sidebar-prioritizer
3,862Relabel and reorder Codex sidebar tasks with P0 to P4 priorities, work categories and B7 age markers, based on evidence from task history.
Trial Reel Distribution
3,862Run trial reels as a distribution lane: repost your best feed videos unchanged to non-followers and graduate the winners.
Video Analysis
3,862Answer questions about a video from its transcript first, and inspect footage with Gemini only when visuals or timing matter.
X Long-Form Post
3,862Write X articles in a founder's voice, with ASCII diagrams and a 24-pattern checklist that catches AI-sounding prose before you post.
graph-colorize
3,792Color-codes the Obsidian graph view by tag, folder, visibility or your own mapping by rewriting only the colorGroups of graph.json, after a backup.
llm-wiki
3,792The foundation skill of obsidian-wiki: a three-layer schema for an LLM-maintained Obsidian vault, with page templates, provenance, confidence, lifecycle and config rules.
memory-bridge
3,792Browse, search and compare Obsidian wiki pages by the AI tool that produced them, and see which tools hold knowledge the others have never touched.
obsidian-layout-adjustment
3,792Restyles Obsidian's own interface with CSS snippets: it maps what you point at to the right UI layer, checkpoints, edits, and verifies each change by screenshot.
vault-skill-factory
3,792Generates a portable, reviewable agent skill from a cluster of mature wiki pages or from one narrated recording of a job, and stops before installing it.
wiki-export
3,792Exports an Obsidian wiki's wikilink graph to JSON, GraphML, Neo4j Cypher, Postgres SQL and an interactive HTML view, plus an optional OKF markdown bundle.
wiki-history-ingest
3,792A router for agent-history ingestion: it picks the Claude, Copilot, Codex, Hermes, OpenClaw or Pi ingest skill from a subcommand or a history path.
li-audit
3,760A post-mortem for your LinkedIn account: ranks past posts by engagement rate and reach multiple, finds the pattern and says what to stop doing.
li-comment
3,760Writes comments on other people's LinkedIn posts that read as a person with an opinion, from nine comment types, two options at a time.