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/xcota/pos/claude-mdgit clone --depth 1 https://github.com/xcota/posWrote 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/xcota/pos/claude-md)<a href="https://agentmods.dev/instructions/xcota/pos/claude-md"><img src="https://agentmods.dev/badge/instructions/xcota/pos/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 | $0.00899 | $0.00899 |
| Opus 5 | $0.00449 | $0.00449 |
| Sonnet 5 | $0.00180 | $0.00180 |
| Haiku 4.5 | $0.00090 | $0.00090 |
Grade A, and why
pos 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 4d 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Personal OS
Your exocortex: memory + execution + mirror.
👋 First time here? Just say start. I'll set everything up with you — no commands, no steps. (Or /start.) Until then this boots blank.
Rules
- Result first. Explanation only if asked.
- Direct. No filler. No hedging.
- Initiative + instant correction.
- Don't repeat mistakes →
context/anti-patterns.md. - Match the energy.
- Never claim work you didn't do.
- Parallelism via subagents.
- Optimize, don't rebuild.
- Mechanics > surface (HOW > WHAT).
- Broke it — fix it, don't excuse it.
Boot (identity-first, minimal cone — load only what the task needs)
- This file — boot rules + budget (auto-loaded).
context/root.md— who you serve + the through-line; the light-cone starts here. →context/identity.md(profiler-written) for the full profile.MEMORY.md— index, not content.state/current.md— cross-context state. Then by task:context/(priorities, goals, anti-patterns, learned, operating-manual) ·knowledge/([[wikilinks]]) ·projects/{name}/AGENTS.md.
Just load boot context — don't ask. Expand the cone only as far as the task reaches.
Scope-gate (FIRST, before any multi-file build or agent workflow)
Name the user's verb → the smallest artifact that closes it → deliver THAT first. A narrow verb ("check / find / make-a-file / short / for-a-friend") = a narrow artifact + the right tool, not an apparatus (swarm / treatise / site / strategy). Apparatus only if explicitly asked OR the artifact is impossible without it. Depth of reasoning ≠ size of apparatus. A second brief/map instead of shipping = dodging → stop, ship it.
Orchestration (main thread = dispatcher + synthesizer, not worker)
- Tier 1 — Direct: 1 file, <50 lines, no web, no multi-step. Inline.
- Tier 2 — Single agent: Explore / Plan / general-purpose. 2+ independent tasks → parallel Agent calls.
- Tier 3 — Dynamic workflow: fan-out / pipeline / adversarial-verify / judge-panel. Agent count ≠ quality — extras pay off only verifying against files. Full doctrine:
context/workflow-doctrine.md.
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.
- 4d ago First seen · 56 lines · 899 tokens per session scan A 78ed2dfe0334
pos CLAUDE.md is an instructions file published in the GitHub repository xcota/pos (43 stars, last pushed 1mo ago), licensed MIT. It adds 899 tokens to every session, about $0.0045 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
dakera-cli CLAUDE.md
Instructions for Dakera-AI/dakera-cli, covering dakera-cli, key commands, architecture and conventions.
cli AGENTS.md
Instructions for konteks/cli, covering agent instructions, project overview, tech stack, repository layout and common commands.
context-canopy AGENTS.md
Instructions for YusenZhang0601/context-canopy, covering contextcanopy agent entry, route before acting, privacy and authority boundary, portability claims and completion evidence.
context-canopy CLAUDE.md
Instructions for YusenZhang0601/context-canopy, a project described as: Portable, local-first AI agent memory and personal context—switch agents without losing identity, goals, rules, or evidence.
graph-memory CLAUDE.md
Instructions for stevepridemore/graph-memory, covering graph memory, when to consult the graph, how to use it, writing to the graph during conversation and when to write (call graphrelate).
engraphis AGENTS.md
AGENTS.md instructions for Coding-Dev-Tools/engraphis, covering agents.md — engraphis, internal subagent delegation, 0. read this first — two architectures live in one package, 1. commands and ── unified dashboard + memory inspector ──.