caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Personal AI assistant — knowledge, tasks, code graph, session intelligence
Plugin anton-core from marketplace anton-core in to-infinity-labs/anton-core-plugin. Commands read from the repo's manifest — yesterday.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
bulk-importImports a whole directory of files into Anton Core's memory, resuming an interrupted run. Use for 'import every markdown file in a folder', 'bulk import this folder' or 'ingest this directory'. For one file, use `save`.
calleesLists every function a function or method calls, directly or transitively, from the code graph. Use for 'what does X call', 'what does X call all the way down', 'what does X depend on' or 'downstream of X', instead of grep or recall.
callersLists every function that calls a function or method, directly or transitively, from the code graph. Use for 'which functions call X', 'who calls X', 'find usages of X' or 'where is X called from', instead of grep or recall.
cyclesFinds circular dependencies and recursive call loops in the code graph, for a package or the whole repo. Use for 'are there circular dependencies in X', 'dependency cycles', 'import cycle' or 'does X call itself'.
dashboardOpens, reuses or stops the read-only browser dashboard. Use for 'open the dashboard', 'dashboard', 'show me the graph/memory/tasks in the browser' or 'stop the dashboard'.
expandShows the full content of saved memories by id. Use for 'show me the full content of item ref-…', 'expand that id', 'tell me more about it', or reading a recall hit before quoting it. Pairs with `explore`.
exploreWalks the relationship graph around one saved memory by id — everything linked to it and how. Use for 'what is connected to item ref-…', 'walk its neighbourhood' or 'what is related to this note'. Pairs with `expand`.
extractPulls action items, decisions, key points and open questions out of meeting notes, a transcript or a document. Use for 'pull the action items out of these notes', 'what did we decide' or 'extract tasks from this'.
healthChecks whether Anton Core is working — memory store, code graph, search, extractor, infrastructure — and names the fix for anything broken. Use for 'run a health check', 'is anything broken', 'is it working' or after setup.
impactShows the blast radius of changing a function, method or type — every caller, subclass and renderer depending on it — from the code graph. Use for 'if I change X, what breaks', 'impact of X' or before a refactor, instead of grep or recall.
improvementsReviews deep analysis's improvement suggestions — lists, approves or dismisses them. Use for 'what did deep analysis suggest I improve', 'pending improvements' or 'review improvements'.
maintenanceRuns maintenance jobs in the foreground — daily cycle, consolidate, calibrate, dedup, purge, prune, retry, reset, status. Use for "run maintenance", "retry failed extractions", "dedupe tasks", "purge stale items", or "check maintenance status".
pathsShows the call chain from one function to another — how A reaches B — from the code graph. Use for 'how does A reach B', 'show me the call chain from A to B' or 'path from main to X'.
recallSearches what Anton Core knows — saved notes, decisions, documents, past sessions, and with --code where a code symbol is defined. Use for 'what do we know about X', 'what did I save about X', 'where is X defined', or loading context before non-trivial work.
relateLinks two saved memories by id — supersedes, resolves, part of, relates to — or suppresses a wrong link. Use for 'record that one note supersedes another', 'link these two notes', 'this resolves that' or 'that link is wrong'.
removeDeletes one saved memory for good, found by id, title or keyword. Use for 'forget the note titled X', 'forget this', 'delete item ref-…' or 'remove that from memory'.
saveSaves a durable fact, decision, preference or correction to Anton Core's memory, in place of native auto memory. Use for 'remember that…', 'remember this', 'save this', 'note that we decided…' or 'store this'.
sessionsReviews how past coding sessions went — counts, durations, tools used, flags, tokens. Use for 'how did my coding sessions go this week', 'session stats', 'session history' or 'what tools did I use'.
setupManages the anton-core install lifecycle, probing state before it acts. Use when the operator wants to install, check, update, repair, reconfigure, or uninstall anton-core, or after a plugin update ships a newer routing fragment.
sharePackages this conversation's work into one self-contained markdown document, saved and copied to the clipboard. Use for 'share this', 'make this portable', 'write this up for someone' or 'package this up'.
summaryBuilds the operator's daily or weekly briefing — open, overdue and due tasks, recent work, new knowledge, spend. Use for 'give me my daily briefing', 'what's on my plate', 'what's due' or 'weekly recap'.
tasksAdds, lists, completes and updates the operator's own to-dos and reminders, kept in Anton Core across sessions. Use for 'remind me to … next Friday', 'add a task', 'I need to …', 'what tasks do I have open' or 'mark that done'. These are the operator's tasks, not this session's work list, and a remi
updateCorrects a saved memory in place by id — title, summary, content or tags. Use for 'that note is wrong, fix it', 'update the memory about X', 'correct item ref-…', 'retag it', or a stale recall hit, instead of saving a second version.
usageReports model-token spend — totals, dollar estimates, per-model and per-project splits, plan headroom — and audits a session's numbers. Use for 'how much have I spent on tokens this week', 'what did that cost', 'token usage' or 'how much headroom is left'.
bash ${CLAUDE_PLUGIN_ROOT}/scripts/session-start.sh --harness claude
bash ${CLAUDE_PLUGIN_ROOT}/scripts/session-end.sh --harness claude
bash ${CLAUDE_PLUGIN_ROOT}/scripts/pre-compact.sh --harness claude
bash ${CLAUDE_PLUGIN_ROOT}/scripts/prompt-submit.sh --harness claude
bash ${CLAUDE_PLUGIN_ROOT}/scripts/pre-tool-use.sh --harness claude
bash ${CLAUDE_PLUGIN_ROOT}/scripts/post-tool-use.sh --harness claude
bash ${CLAUDE_PLUGIN_ROOT}/scripts/post-tool-use-failure.sh --harness claude
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.