This skill should be used when the user says "run khef" on a fresh clone of khef-mac (or asks to "bootstrap khef", "set up khef", "start khef for the first time", "first-time setup khef"). Walks SETUP.md step by step — prerequisites (Docker, Node, ports), JS + Python deps, database (containers, migrations, seeds), MCP…
This skill should be used when the user says "assume identity", "assume the identity", "become ", "this session is the new ", "take over the identity", or wants the current session to become the holder of a named session identity.
This skill should be used when the user says "change name", "rename session", "change my name", "change your name", "rename yourself", or wants to change the current session's nickname.
This skill should be used for ANY kdag, job definition, or pipeline question — including "create a kdag job", "run a kdag job", "build a pipeline", "create a job definition", "new kdag definition", "how does the executor work", "step timeout", "mapreduce behavior", "synthesis", "batchdelayms", "retry semantics", "job…
This skill should be used when the user asks to "create a skill", "add a skill", "new skill", "make a skill", "build a skill", or needs to scaffold a Claude Code skill with proper structure, frontmatter, and placement.
This skill should be used when the user asks to "save this query", "create a saved query", "make a dbx query", "add a saved query for {topic}", "snapshot this saved query", "restore a saved query", or wants to author/run/manage parameterized SQL the agent can re-run later. Also triggers on bare "dbx", "saved query"…
This skill should be used when the user asks to "create a diagram", "draw an ERD", "make a flowchart", "visualize architecture", "diagram this", or needs help creating Mermaid, D2, PlantUML, or Graphviz diagrams with colors and styling.
Extract and view frames from a video file using ffmpeg. Use when the user says "extract frames", "see frames from video", "export video frames", "get screenshots from video", "look at this video", or provides a video file path and wants you to see its contents.
This skill should be used when the user asks for the "session summary", "get session summary", "show the summary", "/get-session-summary", or wants the full summary snapshot of the current session (or another session by nickname). This is the detailed snapshot shown under the "Summaries" tab on the session Summaries…
This skill should be used when the user asks to "test an API endpoint", "create a kapi folder", "add a kapi request", "build a request for {service} API", "scaffold a JWT signer", "save this curl", "run a request against {host}", or wants to check an HTTP call without leaving khef. Also triggers on bare "kapi"…
This skill should be used when the user asks to "add a hook", "create a hook", "manage hooks", "update a hook", "new PreToolUse hook", or needs to create, edit, or install Claude Code hooks in the khef system.
This skill should be used when the user asks to "create a ticket", "add to board", "manage tickets", "create a collection board", "set up a sprint board", or needs to work with khef collections, board views, and ticket-type memories.
This skill should be used when the user asks to "create a metrics dashboard", "add a prometheus widget", "monitor this service", "build an observability dashboard", or needs to create live metrics widgets that query Prometheus and render in khef.
This skill should be used when the user says "onboard", "onboard me on", "catch me up on", "brief me on sessions", or wants to read context from one or more past sessions by nickname to get up to speed.
This skill should be used when the user says "rem-exec", "execute rem report", "apply rem recommendations", or provides a REM report memory UUID to interactively review and execute consolidation recommendations.
This skill should be used when the user says "rem-summary", "curate this summary", "prune the session summary", "edit the session summary", "clean up the summary", or wants to revise a session's current summary snapshot — preserving substantive context (file paths, decisions, errors, open work) while removing process…
This skill should be used when the user says "rem", "consolidate memories", "tidy up", "clean up memories", "memory hygiene", "prune snapshots", "review project knowledge", or needs to scan and report on stale, duplicate, or outdated memories, collections, snapshots, orphaned files, and project knowledge.
This skill should be used when the user says "sss", "search khef", "search source code", "search commits", "search memories", "search sessions", "find in code", "find in khef", or needs to search across khef's indexed knowledge (source code, commits, memories, sessions, slack).
This skill should be used when the user asks to "sync slack", "register a slack channel", "export slack messages", "ingest slack", "search slack", or needs to set up or run the Slack export pipeline.
This skill should be used when the user says "summarize sessions", "add session summaries", "label sessions", "name sessions", or needs to generate short summary labels for sessions that are missing them. Also triggers for "rem sessions".
★not rated 2 15d agoA55 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: