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/indexnetwork/index/claude-mdgit clone --depth 1 https://github.com/indexnetwork/indexWhat 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 | $0.03811 | $0.03811 |
| Opus 5 | $0.01906 | $0.01906 |
| Sonnet 5 | $0.00762 | $0.00762 |
| Haiku 4.5 | $0.00381 | $0.00381 |
Grade A, and why
index CLAUDE.md scanned grade A with 0 findings 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 250 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
How we build: startup rules
This is an early-stage product. Everything in it is provisional and gets rewritten as we learn; nothing in the codebase is owed a future. The git history is full of features that were built with care and then deleted wholesale, plus a long tail of shims, adapters, fallbacks and "legacy" paths that cost effort twice — once to write, once to remove. These rules exist to stop that loop.
- Smallest thing that does what was asked. One code path. No option, mode, flag, fallback or config for a case nobody has today. No abstraction until a second real caller exists. No handling for states the code cannot reach. Test: would a senior engineer call this overcomplicated? If yes, simplify.
- Every changed line traces to the request. No drive-by refactors, renames, reformatting or "while I'm here" improvements. If you notice something worth doing, say so in a sentence; don't do it.
- Change in place, don't layer. When behaviour changes, rewrite the old code: no
legacy*/v2twins, deprecated re-exports, adapters or "compat" paths. Update every caller in this repo in the same PR and delete what the change made dead, tests included. - Backwards compatibility only when it is free. If keeping old data, APIs or behaviour working costs real work — a migration path, dual reads, a parallel code path — don't. Break it, state the break in the PR, move on. This applies to handoffs too: a root session does not write "preserve current behaviour" into a task unless that is the actual goal.
- Don't build for later. No "to make X easier next time". Write the idea down in one line and build it when someone asks for it. YAGNI is the default, not the exception.
- Prefer deleting to maintaining. Unused or redundant code, tests, docs and fixtures get removed, not kept around "just in case".
If the task as written seems to need more than this, say so in a sentence and deliver the small version.
Where things live
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 · 250 lines · 3,811 tokens per session scan A 6dca3f25a2e3
index CLAUDE.md is an instructions file published in the GitHub repository indexnetwork/index (132 stars, last pushed 3d ago), licensed MIT. It adds 3,811 tokens to every session, about $0.0191 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
langgraph AGENTS.md
AGENTS.md instructions for langchain-ai/langgraph, covering agents instructions, corridor security analysis, libraries and dependency map.
agentgateway copilot-instructions.md
Copilot instructions for agentgateway/agentgateway: Do not check for, speculate about, or report compilation errors during code review. Compilation diagnostics from review are frequently incorrect; rely on CI to detect and report compilation failures.
hatch3r CLAUDE.md
Instructions for hatch3r/hatch3r, covering hatch3r — development instructions, architecture, development commands, two-axis pillar framework (2.0.0) and orchestrator self-discipline (bypass protection).
aeon CLAUDE.md
Instructions for aeonfun/aeon, covering aeon, how aeon works, strategy, voice and soul file hierarchy (read in this order).
ken CLAUDE.md
Instructions for townsendmerino/ken, covering claude.md, what this is, repository ownership (read this first), commands and embedding parity & golden fixtures (now in aikit).
gangsta GEMINI.md
Instructions for kucherenko/gangsta, a project described as: AI agentic skills framework for spec-driven development, built on the organizational model of mafia.