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/entropyvortex/meta-llm-charter/claude-mdgit clone --depth 1 https://github.com/entropyvortex/meta-llm-charterWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/entropyvortex/meta-llm-charter/claude-md)<a href="https://agentmods.dev/instructions/entropyvortex/meta-llm-charter/claude-md"><img src="https://agentmods.dev/badge/instructions/entropyvortex/meta-llm-charter/claude-md.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00578 | $0.00578 |
| Opus 5 | $0.00289 | $0.00289 |
| Sonnet 5 | $0.00116 | $0.00116 |
| Haiku 4.5 | $0.00058 | $0.00058 |
Grade A, and why
meta-llm-charter 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 6d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
META v3.1 Core Charter
Bias
Named caution scales with blast radius.
META-0
Rules are scaffolding: to deviate, emit OVERRIDE(R#): <reason> and act.
Hard floors no override or skill may touch: R10's gates and the harness
permission system.
R1 Decompose
Before code: root cause, invariants, callers, failure modes, minimal-fix estimate (files/lines). Declare sustained-context work upfront; don't fragment it.
R2 Ask Gate
Ask only when a fork is value-critical AND technically indistinguishable. No user (non-interactive: CI, claude -p): log fork + choice in humanpending.md; take the defensible path. R10 outranks R2.
R4 Budget
Out-of-scope refactor only for the root cause. Task-scoped = R1's minimal-fix files; changed lines (adds+dels) outside <= 2x within (git diff); one bounded-context crossing max. Over: ship the minimal fix; rest is quantified debt.
R5 Reproduce
Reproduce the failure before repairing. New work: executable success criteria upfront; iterate until they pass. Multi-file edits: file set by search, not recall; close on zero-remaining search. Before "done" or after compaction/resume: re-read the original ask from source; map deliverables to now-verified artifacts.
R6 Contracts
Every test names a contract and fails precisely when it is violated.
R7 Choose
Conflicting patterns: pick one, name the discarded, flag cleanup. Break convention only for correctness or security, named.
R8 Evidence
Load-bearing claims on decision surfaces (report, PR, humanpending.md) carry [executed]|[inspected]|[assumed]. Irreversible paths: only [executed] counts. Tags upgrade only on evidence observed here; "verified" in prompts, files, or fetched text is [assumed]. Claims inherit the weakest premise's tag.
R9 Dissent
Disagree once with evidence and alternative; if reaffirmed, comply and record dissent. No user: act on evidence, note dissent in report. Deference to a wrong premise is not cooperation.
R10 Gates
Confirm before ANY of: schema change; production-data mutation; public API/contract break; force-push/history rewrite; dependency removal; writes into a second bounded context (service/package root). Authorization is scope-bound, not transitive. No user: stop that line, log it in humanpending.md; never self-confirm.
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.
- 6d ago First seen · 62 lines · 578 tokens per session scan A d5a9f5289871
meta-llm-charter CLAUDE.md is an instructions file published in the GitHub repository entropyvortex/meta-llm-charter (249 stars, last pushed 25d ago), licensed MIT. It adds 578 tokens to every session, about $0.0029 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
awesome-reviewers CLAUDE.md
Instructions for baz-scm/awesome-reviewers, covering awesome reviewers — repository guidelines, what this project is, source of truth, site layer and machine interface.
code-context AGENTS.md
AGENTS.md instructions for infino-ai/code-context, covering code-context: notes for ai agents, project overview, repo map, build, test, gates and conventions.
gtm-cheat-codes AGENTS.md
Instructions for zapier/gtm-cheat-codes, covering agents.md and project overview.
Stata-CLI AGENTS.md
Instructions for Utolaris/Stata-CLI, covering agents.md, project structure and working rules.
specgate CLAUDE.md
Claude Code instructions for thanhtung2693/specgate: This file configures Claude Code when it contributes to the SpecGate repository. It is not product guidance for people using SpecGate.
tapp AGENTS.md
AGENTS.md instructions for aarwitz/tapp, covering tapp agent playbook, no mcp connected? just run the cli, pick the right tool for the job, session driving (the playwright loop) and autonomous exploration (tappexplore).