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/kleosr/kleosrules/agents-mdgit clone --depth 1 https://github.com/kleosr/kleosrulesWhat 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 | $0.00417 | $0.00417 |
| Opus 5 | $0.00209 | $0.00209 |
| Sonnet 5 | $0.00083 | $0.00083 |
| Haiku 4.5 | $0.00042 | $0.00042 |
Grade A, and why
kleosrules 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 yesterday.
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.
What it actually says
AGENTS.md map of this pack
LAW vs MAP
- Law: rules paste + rules + registered hook scripts
- This file: navigation only
Overview kleosrules V2 Bash hooks + local NOW.md memory. Brain equals NOW.md (local). Muscle equals four registered hook scripts. No Rust. No Python pack tooling. No MCP core dependency.
Install scope
Local install is global-only: FORCE=1 bash scripts/install.sh writes ~/.cursor (hooks, GLOBAL rules, skills, hunter/cut/prove). sync is opt-in via empty-by-default shared/config/scan.roots. Never install Lane-A into this pack.
Where to look
- Paste: shared/rules/USER-RULES.paste.txt
- Hooks: shared/hooks/session_start.sh before_submit_prompt.sh before_shell.sh before_read_file.sh
- Hook lib: shared/hooks/lib/common.sh shell_gate.sh shell_fleet.sh
- Install: shared/hooks/fleet_sync.sh install → ~/.cursor
- Agents: shared/agents/hunter.md cut.md prove.md
- Platform installers: MacOS/install.sh Linux/install.sh Windows/install.ps1 (+ Windows/hooks/wsl-shim.ps1)
- Scripts: scripts/doctor.sh scripts/install.sh scripts/sync.sh
- Tests: tests/run.sh + tests/fixtures/
- Architecture: docs/ARCHITECTURE.md
- Curator: docs/CURATOR.md
- Verify: docs/TOOLCHAIN.md
- NOW.md: bounded session state (compaction protocol)
- SECURITY.md: pnpm + cybersecurity SSOT
Done chmod +x shared/hooks scripts then bash -n on hook scripts + bash scripts/doctor.sh
Hard stops
- Never reintroduce Rust kleos-gate or pack Python
- Never emit updated_input (this pack does not register preToolUse); inject via additional_context (sessionStart) or continue (beforeSubmitPrompt)
- Secrets never in paste hooks or chat
- Never make MCP a core dependency (optional only)
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.
- yesterday First seen · 37 lines · 417 tokens per session scan A 53db69b9bb32
kleosrules AGENTS.md is an instructions file published in the GitHub repository kleosr/kleosrules (2 stars, last pushed yesterday), licensed MIT. It adds 417 tokens to every session, about $0.0021 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 2026-08-31.
Other instructions, from other repositories
cursor-os AGENTS.md
Instructions for KingEmma7/cursor-os, covering agents.md — cursor-os, what this project is, repository layout, working agreements and verification before done.
syncai copilot-instructions.md
Instructions for dmtrkzntsv/syncai, covering claude.md, build & run, architecture, lifecycle and identify is the routing decision.
cursor-rules AGENTS.md
Instructions for ZanzyTHEbar/cursor-rules, covering agents.md, repo shape, commands, config and runtime gotchas and tests and fixtures.
agent-hook-schemas CLAUDE.md
Claude Code instructions for mherod/agent-hook-schemas, covering claude.md, build & test commands, npm publish, architecture and key patterns.
uikit-expert-skill AGENTS.md
Instructions for zinxj/uikit-expert-skill, covering agent guidelines for uikit expert skill, core principles, 1. uikit focus only, 2. no code formatting or linting and 3. no architectural opinions.
post_compact_reminder AGENTS.md
AGENTS.md instructions for Dicklesworthstone/post_compact_reminder, covering agents.md — post-compact reminder for claude code, rule 0 - the fundamental override prerogative, rule number 1: no file deletion, git branch: only use main, never master and toolchain: bash & shell scripts.