Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/agentidx/agentindex/claude-mdgit clone --depth 1 https://github.com/agentidx/agentindexWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.03519 | $0.03519 |
| Opus 5 | $0.01759 | $0.01759 |
| Sonnet 5 | $0.00704 | $0.00704 |
| Haiku 4.5 | $0.00352 | $0.00352 |
Grade B, and why
agentindex CLAUDE.md scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
6. **Pending sudo fixes:** `scripts/apply_system_limits.sh`, auto-login anstudio, UPS purchase. Sudo password unknown -- needs recovery. How it starts
The opening of the file, as written. The whole thing — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md -- Orientation for Claude sessions working on Nerq/ZARQ
Read this file first. It is the map to everything else.
Then read ~/agentindex/PROGRAM.md — the operating board: everything running (timers+guards), in flight, staged, held on Anders, the decision log, and the north-star values. It is the single source of truth for program state.
STANDING RULE — the board is not optional. Every Code session MUST end by updating PROGRAM.md: move status (RUNNING/IN PROGRESS/STAGED/HELD/FROZEN-TEST/DONE), log any signed decision in the Decision Log with today's date, reorder the staged queue, refresh north-star values, and bump the "Last board update" date. A session that changed program state but didn't update the board isn't finished. The heartbeat flags the board as a silent-death class if it goes >8 days untouched.
If there is a docs/session-handoff-<date>.md file newer than this one, read that first — it contains the most recent session's context and specific tasks for you to pick up.
The critical fact you must know before touching anything
Nerq is operated 24/7 by an autonomous AI agent named Buzz, running inside openclaw (~/.openclaw/). Buzz was configured by a previous Claude instance. The system has three participants:
- Anders Nilsson -- strategic decisions
- Buzz -- 24/7 operations, health monitoring, self-healing
- Claude (you) -- advisory, building, fixing
When things happen in the system that neither you nor Anders did, it is probably Buzz. Do not assume you know what Buzz is doing without reading its config.
Required reading before operational work: docs/buzz-context.md
What is this system?
- ZARQ (zarq.ai) is a crypto risk intelligence platform -- "Moody's for crypto". Trust Score, Distance-to-Default, crash probability.
- Nerq (nerq.ai) is an AI asset search engine indexing 5M+ AI assets (204K agents & tools, 4.7M models & datasets) with Trust Scores.
- Both run on the same FastAPI backend on one Mac Studio M1 Ultra (64 GB RAM).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 186 lines · 3,519 tokens per session scan B 3b158aa984ce
agentindex CLAUDE.md is an instructions file published in the GitHub repository agentidx/agentindex (2 stars, last pushed 1mo ago), licensed MIT. It adds 3,519 tokens to every session, about $0.0176 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
OwnPilot AGENTS.md
Instructions for ownpilot/OwnPilot, covering ownpilot, architecture, key patterns, commands and tech stack.
agentic-os AGENTS.md
Instructions for modimihir07/agentic-os, covering agentic os — complete project context for ai agents, role definition, project identity, architecture and 3-agent engine.
zeroshot AGENTS.md
AGENTS.md instructions for the-open-engine/zeroshot, covering zeroshot: multi-agent coordination engine, critical rules, where to look, cluster worker contracts and cli quick reference.
CORE CLAUDE.md
Instructions for DariuszNewecki/CORE, covering claude.md — core, what core is, and what you are, source layout, how to work in this repo and governed and prohibited surfaces.
NEEDLE CLAUDE.md
Instructions for jedarden/NEEDLE, covering needle project conventions, msrv, module dependency graph, code style and testing.
nightshift AGENTS.md
Instructions for orwa-mahmoud/nightshift: AI-assisted contributions are welcome. The same correctness, security, licensing, testing, and quality standards apply regardless of which tools are used.