tonydzi

66 mods across 11 repositories, 37 stars between them.

tonydzi/claude-bible

Instructions file CodexOpenCode

Instructions for tonydzi/claude-bible, covering agents.md — working in this repo, what this repo is, layout, how to verify a change and conventions.

9 2d ago A 864 tokens original MIT

tonydzi/verbatim-citation-gate

Instructions file CodexOpenCode

Instructions for tonydzi/verbatim-citation-gate, covering agents.md — working in this repo, what this repo is, stack and layout, how to verify a change and conventions.

3 2d ago A 846 tokens original MIT

tonydzi/sqlite-graph-memory

Instructions file CodexOpenCode

Instructions for tonydzi/sqlite-graph-memory, covering agents.md — working in this repo, what this repo is, stack and layout, how to verify a change and conventions.

3 2d ago A 939 tokens original MIT

tonydzi/claw-consensus

Instructions file CodexOpenCode

Instructions for tonydzi/claw-consensus, covering agents.md — working in this repo, what this repo is, stack and layout, how to verify a change and boundaries — what needs a human.

2 2d ago A 979 tokens original MIT

charm-os AGENTS.md

05

tonydzi/charm-os

Instructions file CodexOpenCode

Instructions for tonydzi/charm-os, covering agents.md — working in this repo, what this repo is, stack and layout, how to verify a change and conventions.

2 3d ago A 895 tokens original Apache-2.0

tonydzi/agent-leash

Instructions file CodexOpenCode

Instructions for tonydzi/agent-leash, covering agents.md — working in this repo, what this repo is, layout, how to verify a change and conventions — the claim discipline.

2 2d ago A 834 tokens original MIT

second-brain

07

tonydzi/second-brain-starter-kit

Plugin Claude Code

Battle-tested Claude Code skills from a live second-brain and AI-cofounder operation at Palo Alto AI Research Lab: quality gates, retrospectives, memory and recall, fleet coordination. Method, not data.

1 2d ago A tokens not measured original MIT

second-brain-skills

08

tonydzi/second-brain-starter-kit

Plugin Claude Code

101 battle-tested Claude Code skills from a live second-brain + assistant operation. Quality gates, retrospectives, CRM, vault ingest, multi-model review panels, fleet coordination. Method, not data.

1 2d ago A tokens not measured original MIT

3

10

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Launch autonomous multi-machine consensus between Claude peers (hub + laptops + teammate machines): the peers negotiate a decision among themselves over a shared coordination channel and execute it, without using the human as a courier. Trigger on "/03" or phrases like "work it out among yourselves" / "find…

1 2d ago A 97 tokens original MIT

1

11

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Recover after a HARD session crash (app died, computer rebooted, context lost mid-work) with one ultra-short command: where were we + is everything alive + restore history. Trigger on "/1", "/!", "resume after crash", "wake". Three steps: (1) RECALL from the TurnState black-box ledger (last requests, files, decisions…

1 2d ago A 130 tokens original MIT

agenda

12

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

The operator's day at a glance, on demand — today's calendar plus what needs THEM personally (leads due, real humans awaiting a reply, deadlines) pulled from Calendar + Telegram, ranked. Trigger on "/agenda", "what's on today", "my agenda". READ-ONLY summary — it never sends or commits anything. Pairs with a…

1 2d ago A 83 tokens original MIT

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Mandatory decision protocol ("Alpha Protocol") for any new STRATEGIC work — never jump to implementation on local recall alone. Run RECALL over the knowledge vault → GAP analysis → emit a DEEP RESEARCH PROMPT to run in external deep-research tools → SYNTHESIS + Decision Memo when results come back. Trigger on…

1 2d ago A 129 tokens original MIT

alpha-judge

14

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

The LLM-JUDGE stage of an alpha-extraction engine — the reusable "expensive judge" half of the "cheap detector → expensive judge" pattern, shared by all miners. Given a miner's deterministic candidate list (0-token prefilter), it keeps only the REAL signal and drops noise, then proposes additions to that miner's…

1 2d ago A 108 tokens original MIT

alpha-review

15

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Open the ALPHA REVIEW screen — the one place to mark judge keepers as gold/miss and see per- miner precision — and print the eval state. Trigger on "/alpha-review", "alpha screen", "open the review screen". Thin launcher over the existing harvest engine and local review server; builds nothing new.

1 2d ago A 67 tokens original MIT

arch

16

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Read the "System Architect" map — a deterministic catalog of EVERYTHING the system is made of (vault + import scripts + scheduled tasks + MCP servers + SQLite DBs + hooks + skills + dashboards), what's healthy, and what fell off. Trigger on "/arch", "/arch broken", "/arch scan", "system health", "what broke in the…

1 2d ago C 86 tokens original MIT

ask

17

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Ask the second brain in plain words — semantic search over the curated knowledge vault (embeddings + reranker), not a full-corpus dump. Trigger on "/ask ", "what do I have on ", "what did I write about X". Token-cheap: RAG retrieves the smallest relevant slice, the LLM only synthesizes the top hits.

1 2d ago A 77 tokens copy · 88% MIT

bible

18

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

The operations Bible — a single behavioral codex governing everyone who acts AS or FOR the owner: the owner, human assistants, and AI agents. Load BEFORE doing or saying anything on the owner's behalf — outreach, replying in their chats, scheduling, purchases, hiring, household ops. Explains where the rules live (the…

1 2d ago A 92 tokens original MIT

brain

19

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

One-glance health of the "second brain" so failures aren't SILENT — is the local search server alive, is the RAG index fresh, is the session-memory ledger filling, did the nightly distillation run. Trigger on "/brain", "/memory", "brain health", "is the reindex alive". READ-ONLY diagnostic with a green/red verdict per…

1 2d ago A 77 tokens original MIT

bus

20

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Cross-machine messaging between computers in the fleet over a shared TELEGRAM GROUP (primary rail) — every machine posts/reads in one group via the Telegram MCP, so it works even when file sync is down, AND humans can watch the machines talk in plain sight. A synced folder stays as the FALLBACK for oversized payloads.…

1 2d ago A 90 tokens original MIT

canon-revision

21

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Full structural REVISION of an always-loaded rules file (CLAUDE.md / MEMORY.md): build a TOP block of the most important rules + numbered categories, move rule bodies verbatim (a script moves bytes and proves 0 losses; the LLM decides only the mapping), fold heavy mechanics into trigger+gist+pointer form, then guards…

1 2d ago A 104 tokens original MIT

cc

22

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Hotkey alias for "compact the context": prints a READY-made /compact line in a canonical handoff format so the operator can paste it as the next message and squeeze working memory without losing the thread. Trigger on "/cc", "compact", "finish via compact". Part of the short-alias family (/tt /rr /cc /1).

1 2d ago B 75 tokens original MIT

chat-search

23

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Search INSIDE past Claude Code sessions (the episodic "book of chats" index) — find which previous session discussed a topic, and continue it with one command. Uses a SEPARATE session index, NOT the curated-knowledge RAG, so raw chat noise never pollutes the distilled brain. Trigger on "/chat-search", "which chat…

1 2d ago A 88 tokens original MIT

chat

24

tonydzi/second-brain-starter-kit

Skill Claude CodeCodex

Find a Telegram CHAT/group/channel by name instantly — returns its telegramid + t.me link from a local chats.db index (0 tokens, no live crawl). Trigger on "/chat ", "find the chat", "chat id for ". Also answers "do we share any groups with " via common-groups lookup. Distinct from person-search: this is the CHAT lane.

1 2d ago A 87 tokens original MIT