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/pratiyush/llm-wiki/agents-mdgit clone --depth 1 https://github.com/Pratiyush/llm-wikiWrote 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/pratiyush/llm-wiki/agents-md)<a href="https://agentmods.dev/instructions/pratiyush/llm-wiki/agents-md"><img src="https://agentmods.dev/badge/instructions/pratiyush/llm-wiki/agents-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 | $0.01722 | $0.01722 |
| Opus 5 | $0.00861 | $0.00861 |
| Sonnet 5 | $0.00344 | $0.00344 |
| Haiku 4.5 | $0.00172 | $0.00172 |
Grade A, and why
llm-wiki AGENTS.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 5d 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 — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
llmwiki — Codex CLI / Agent Schema
This file is the schema for Codex CLI and any other coding agent that reads AGENTS.md instead of CLAUDE.md (OpenCode, Gemini CLI, etc.). The workflows are identical to CLAUDE.md — only the language is agent-agnostic.
Three layers
raw/ IMMUTABLE. Session transcripts converted from the agent's session store.
Never modify files here.
wiki/ LLM-maintained. Pages you write that summarise and cross-reference raw/.
index.md Catalog of every page. Update on every ingest.
log.md Append-only chronological record (auto-archives at 50KB).
overview.md Living synthesis.
hints.md Writing conventions and entity naming rules (load on demand).
hot.md Last 10 session summaries — global hot cache.
hot/<project>.md Per-project hot caches (default ON, configurable).
MEMORY.md Cross-session facts (200-line cap, auto-consolidated).
SOUL.md Wiki identity and writing voice.
CRITICAL_FACTS.md Must-know facts (<120 tokens).
sources/ One page per raw source.
entities/ People, companies, projects, products.
concepts/ Ideas, frameworks, methods.
syntheses/ Saved query answers.
site/ GENERATED. Static HTML from `python3 -m llmwiki build`. Do not edit.
Session stores by agent
Different agents write their transcripts to different locations. The adapter registry in llmwiki/adapters/ abstracts this away.
| Agent | Session store | Adapter |
|---|---|---|
| Claude Code | ~/.claude/projects/<project>/<uuid>.jsonl |
claude_code.py |
| Codex CLI | ~/.codex/sessions/ |
codex_cli.py (stub in v0.1) |
| Gemini CLI | ~/.gemini/ (TBD) |
gemini_cli.py (planned) |
| OpenCode | ~/.opencode/ (TBD) |
opencode.py (planned) |
The CLI auto-detects which adapter(s) to run. Override with --adapter <name>.
Commands
Run from inside the repo:
python3 -m llmwiki sync # convert new .jsonl → raw/sessions/*.md
python3 -m llmwiki build # compile site/ from raw/ + wiki/
python3 -m llmwiki serve # local HTTP server on 127.0.0.1:8765
python3 -m llmwiki init # scaffold raw/, wiki/, site/ directories
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.
- 5d ago First seen · 214 lines · 1,722 tokens per session scan A d60cda1397ec
llm-wiki AGENTS.md is an instructions file published in the GitHub repository Pratiyush/llm-wiki (383 stars, last pushed 2mo ago), licensed MIT. It adds 1,722 tokens to every session, about $0.0086 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
hop CLAUDE.md
Instructions for amittamari/hop, a project described as: Fast full-text search and resume for coding-agent sessions.
obsidian-mind CLAUDE.md
Claude Code instructions for breferrari/obsidian-mind, covering obsidian mind, skills & capabilities, custom slash commands, vault structure and obsidian cli.
repo-context-mcp AGENTS.md
Instructions for nduc99911/repo-context-mcp, covering agents.md — repo-context-mcp, project, layout, commands and conventions.
tages CLAUDE.md
Instructions for ryantlee25-droid/tages, covering tages — project configuration, architecture, key design decisions, security and commands.
pixeltamer-gpt-image-skill CLAUDE.md
Instructions for gabelul/pixeltamer-gpt-image-skill, covering claude.md — pixeltamer, what this is, architecture, how the skill works and the two backends.
toon-memory AGENTS.md
Instructions for LuiggiVal08/toon-memory, covering agents.md, build & verify, architecture, smarter memory pattern and visual memory pattern.