findtorontoevents_antigravity.ca-archive-2026-05-23

Semantic memory layer for long-term knowledge storage and retrieval across sessions. Use for /agentmemory remember, search, context, forget, status commands. Distinct from holographic memory (live cross-agent state) — agentmemory is a searchable knowledge base backed by BM25+vector hybrid search at localhost:3111.

Alias for /consult-ring261T. Consult Ring 2.6 1T (inclusionai/ring-2.6-1t) on OpenRouter. Use when user says "/ask-ring261T" or "ask ring".

Use when tracing how a pick on findtorontoevents.ca/audit got there — what source emitted it, which gates it passed or was rejected by, why it is or is not in Smart Picks / High Conviction / Money Ready, or why a class shows the win-rate it does. Also for "which DB tables matter", per-asset-class pick-funnel case studies, and onboarding someone to the audit pipeline.

Use when user asks to check GitHub Actions for failing jobs, chronic cancellations, or jobs without a subsequent successful run. Also for CI/CD health audits, rerunning failed workflows, or monitoring workflow reliability. Aliases - checkGHA, gha_health, github_actions_health, gh_actions, check_actions, fixjob, github_actions, actions_health

Get a second opinion from OpenAI Codex via its headless CLI. Use when the user says "/consult-codex", "ask codex", "codex second opinion", or wants an outside model's take on a question, plan, or code.

Get a second opinion from Cursor Agent via its headless CLI (the `agent` command). Use when the user says "/consult-cursor-agent", "ask cursor-agent", "cursor-agent second opinion", or wants an outside model's take on a question, plan, or code.

Get a second opinion from Google Gemini via its headless CLI. Use when the user says "/consult-gemini", "ask gemini", "gemini second opinion", or wants an outside model's take on a question, plan, or code.

Get a second opinion from xAI Grok via its headless CLI (`grok -p`). Use when the user says "/consult-grok", "ask grok", "grok second opinion", or wants an outside model's take on a question, plan, or code.

Get a second opinion from kilo (opencode fork) via its headless CLI. Use when the user says "/consult-kilo", "ask kilo", "kilo second opinion", or wants an outside model's take on a question, plan, or code.

Get a second opinion from opencode via its headless CLI. Use when the user says "/consult-opencode", "ask opencode", "opencode second opinion", or wants an outside model's take on a question, plan, or code.

Consult Ring 2.6 1T (inclusionai/ring-2.6-1t) on OpenRouter for deep reasoning on quant, strategy, hypothesis, or code questions. Use when the user says "/consult-ring261T", "/ask-ring261T", "ask ring", "ring second opinion", or wants the 1T-parameter extended-thinking model's take on a question.

Check messages on the cross-PC protocol gateway — poll your peer inbox, drain the broadcast queue (peer_id=all), and tail the durable JSONL event log. Use when the user says "check cross-pc msgs", "any peer messages", "/cross-pc-checkmsg", "what's in the bus", or wants to inspect inbound traffic. Aliases - xpc-check, cross-pc-poll, checkmsg.

Quick health check of the local cross-PC protocol gateway — confirms /health, lists registered peers, queue depths, pending ACKs, LAN-discovered peers. Use when the user says "is the gateway up", "/cross-pc-health", "cross-pc status", or you need a pre-flight before sending. Aliases - xpc-health, cross-pc-status.

Use when implementing, operating, or debugging the cross-PC communication protocol that bridges Cursor, Claude peers, and Freebuff-style CLI agents across machines. Covers WebSocket+HTTP transport, canonical envelope, ACK/retry, store-and-forward, replay, LAN discovery, Redis backward-compat bridge, and the JSONL inspector workflow. Aliases - cross-pc, xpc, protocol-debug, cross_pc_protocol

Send a message over the cross-PC protocol (WebSocket primary, HTTP fallback). Use when the user says "send a cross-pc message", "broadcast to peers", "DM peer X", "/cross-pc-sendmsg", or wants to publish an envelope to the local protocol gateway. Aliases - xpc-send, cross-pc-send, sendmsg.

Look up database schema, table structure, or connection details for any of the 9 MySQL databases (ejaguiar1_stocks, ejaguiar1_backtests, ejaguiar1_sportsbet, ejaguiar1_favcreators, ejaguiar1_events, ejaguiar1_memecoin, ejaguiar1_tvmoviestrailers, ejaguiar1_news, ejaguiar1_deals). Triggers — `/db-schema`, "what tables are in X db", "schema for table Y", "how is the DB structured", "where is the schema", "DB connection details", "regenerate schema doc". Aliases — dbschema, db-schema, schema-lookup, db-lookup.

Push the current session's transcript summary + open PRs + outstanding action items to the cross-PC protocol gateway so peer agents (Cursor, Hermes/WSL, Freebuff, Claude on laptop, etc.) can see what this instance just did. Then poll the broadcast + DM queues for any messages from those peers and surface them to the operator. Optionally revise + recommit work based on peer insights. Triggers — `/dropchat-multipc`, "drop chat to multi-pc", "broadcast session summary", "DROPCHATTOMULTIPC", "share with peers", "check peer inbox + sync". Aliases — dropchat, dctm, multipc-sync, peer-sync.

Use when checking for failed GitHub Actions jobs that have no subsequent successful run, or when user wants to find and rerun stale failures. Aliases - fixGHACTIONS, FIXGH, fixjobs

Set a persistent standing objective that survives across turns and auto-continues until a judge confirms it is done. Use when the user says "/goal", "set a goal", "keep working until X", "work toward X autonomously", or wants Claude to self-continue toward an objective without re-prompting each turn. Subcommands - status, pause, resume, clear. Aliases - goal, standing-objective, autocontinue.

Query the existing Graphify knowledge graph to navigate this codebase fast. Use for "where is X defined", "what calls Y", "how do A and B connect", "explain this module", "what is in graphify". Reads graphify-out/graph.json — no LLM cost. Aliases - graphify-query, ask-graph, code-nav.

Build or rebuild the Graphify knowledge-graph map of this gigantic repo SAFELY. Use when asked to "map the codebase", "graphify the repo", "regenerate the code graph", or "index docs/code into graphify". Encodes the workarounds for this repo's 20K-file scale. Aliases - graphify-build, map-codebase, index-repo.

Incrementally update the Graphify graph after code changes — re-extract only the touched directories and re-merge, instead of a full rebuild. Use after a commit, after editing modules, or when graphify-out is stale. Aliases - graphify-update, refresh-graph, regraph.

any input (code, docs, papers, images, videos) to knowledge graph. Use when user asks any question about a codebase, documents, or project content - especially if graphify-out/ exists, treat the question as a /graphify query.

Autonomously work through every outstanding TODO / action item in the session until done, then sweep the conversation for anything missed. Use when the user says "/handlealltodos", "handle all todos", "finish the backlog", "clear all action items". Wraps the /goal loop with a chat-sweep + escalate-to-AI rule.

Read from and write to the shared Hermes/Grok holographic memory (agent_shared_memory.json). Use when the user says "/holographic-memory", "holo-mem", "check holographic memory", "write to holographic memory", or "what does hermes remember". Manages the cross-session/cross-PC agent continuity layer separate from personal MEMORY.md.

Audit findtorontoevents.ca homepage for performance issues and filter bugs, including date-filter chips (All Dates, Today, This Week, This Month), the Dating/evergreen category, multi-day events toggle, and near-me proximity filter. Covers both the React event-grid widget and the vanilla JS applyFilters() layer in index.html. Aliases - homepageaudit, filteraudit, homepage-check, perf-audit

Use when someone asks what the hypothesis registry is, why it exists, what pre-registration / M-107 means, what a harness or harness run is, or what "eff" / "admissible" mean. Conceptual explainer, not a workflow.

Use when registering a new trading-strategy hypothesis, running the admissibility harness on one, or recording a verdict. Operational workflow for reports/hypothesis_registry.json (rule M-107 — pre-register before any backtest).

Safe, fast git operations for large repositories that time out on naive commands. Covers: status, fetch, diff, log, sync-from-remote, find-files, and avoid-kills. Use when any git operation is slow, killed, or timing out. Aliases - bigrepo, git-fast, repo-sync, git-safe, large-repo

Safe read patterns for large files in this repo — dashboard JSON (18MB), coinglass DB (27MB), strategy JSONs, log files. Prevents OOM crashes and timeout kills from naive cat/read. Aliases - bigfile, safe-read, large-file-read, repo-read

Use when user wants to audit findtorontoevents.ca/audit (+ /audit/hyrotrader, ejaguiar1_stocks, ejaguiar1_backtests) for production-grade real-money readiness. Per-asset-class statistical-edge hunt with cross-source correlation, external-data integration suggestions, dashboard UI/filter audit, and ranked Best-Possible-Action recommendations. Triggers — `/money-maker-ready`, "is /audit ready for real money?", "find edge per asset class", "money maker audit", "edge hunt", "what's our best statistical edge?", "should we go live with X asset class?". Aliases — mmready, money-maker, edge-hunt, real-money-audit, edge-by-class.

Ultimate statistical edge audit per asset class. Produces proven, real-money-grade filters for CRYPTO/EQUITY/COMMODITY/ETF/FOREX/BOND so that a quant or hedge fund manager would find the edge trustworthy. Extends money-maker-ready v1.1 with stronger success criteria, autonomous execution rules, and a weekly filter output. Use when the user says "/money-maker-readyv2", "prove the edge", "real money filters", or "ultimate statistical edge". Aliases - money-maker-readyv2, ultimate-edge, prove-edge.

Generates a no-shortcuts system prompt for querying any AI model, requiring production-grade reasoning, cited sources, no hallucinated data, and complete output. Use when the user says "/noshortcutsprompt", "no shortcuts", or "production grade prompt".

Alias for swarm-transcript-scan — audits a long chat transcript for missed/unfinished action items. Use when the user types /swarm-transcript-review or /swarm-transcript-scan.

Scan a long Claude Code session transcript for action items / TODOs the session may have dropped. Use when the user says "/swarm-transcript-scan", "scan the transcript", "did I miss any action items", "find missed TODOs in this chat", or wants a long chat audited for unfinished work. Wraps tools/swarm/transcript_action_scan.py.

Alias for swarm-transcript-scan — chunks a long Claude Code session transcript and audits it for missed/unfinished action items. Use when the user types /swarm-transcript (with or without a -scan / -review / -summary suffix), "scan the transcript", "did I miss anything in this chat".

Switch between TradingView paper portfolios reliably using DOM index-click (NOT text-match, which has substring collisions on this user's account names). Verifies the post-switch account name, handles the dropdown's between-call auto-close, and warns when the user has manually clicked into a different account. Aliases — tv-switch, tv-account, tv-pick-acct.

Launch TradingView Desktop with Chrome DevTools Protocol so the tradingview-desktop MCP can drive the UI. Handles the WindowsApps install path (which the MCP's tv_launch can't auto-find), the non-default-port problem (MCP defaults to 9222 but TV is often bound to 9223), and the loopback-only listen behavior. Aliases — tv-launch, tv-up, tv-cdp.

Close one or more TradingView paper positions safely. Verifies the account first (drift kills the wrong portfolio), uses row-level Close buttons (NOT the global flat-all), confirms close in the History/Trades log, and handles closed-market behavior (US equity after-hours, futures session gaps). Aliases — tv-close, tv-flat, close-positions.

Diagnostic matrix for every failure mode encountered when driving TradingView Desktop via the tradingview-desktop MCP — CDP unreachable, port-mismatch, dropdown auto-close between evals, History-tab shadowing the positions reader, account drift after user clicks, Unicode minus sign in PnL, BLOCKER_FAIL on TP/SL inputs, async eval returning empty object. Aliases — tv-fix, tv-troubleshoot, tv-doctor.