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/lilmgenius/polysona/claude-mdgit clone --depth 1 https://github.com/LilMGenius/polysonaWrote 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/lilmgenius/polysona/claude-md)<a href="https://agentmods.dev/instructions/lilmgenius/polysona/claude-md"><img src="https://agentmods.dev/badge/instructions/lilmgenius/polysona/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 | $0.00623 | $0.00623 |
| Opus 5 | $0.00311 | $0.00311 |
| Sonnet 5 | $0.00125 | $0.00125 |
| Haiku 4.5 | $0.00062 | $0.00062 |
Grade A, and why
polysona 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Polysona
Build and run multiple personas across any AI agent.
Philosophy (Agent-Enforced)
- Execution velocity — ship visible outcomes quickly and iterate daily.
- Data flywheel — every run should improve persona quality and future outputs.
- Extreme pragmatism — start from minimal working units and avoid overengineering.
- Top-down prioritization — tackle the hardest constraint first, then descend.
- Facts first — verify before deciding; do not guess when data can be checked.
- Open-source ecosystem — build in public for interoperability, trust, and contribution.
- No speculation — uncertain claims are blocked until grounded by evidence.
Behavioral Rules (No Exceptions)
- No speculation: never invent numbers, framework counts, timelines, or capability claims.
- Single Source of Truth (SSOT): define each fact in one owner document; reference it elsewhere.
- Tone consistency: keep style stable in each file; avoid repetition and verbosity.
- Immediate feedback loop: apply user corrections immediately and persist them in the right files.
Context Loading Protocol
Always read required context before generating outputs.
| Trigger | Mandatory Read |
|---|---|
| Interview session | personas/{id}/persona.md, interview skill references, latest interview-log |
| Persona introduction | personas/{id}/persona.md, personas/{id}/nuance.md, personas/{id}/accounts.md |
| Trend detection | relevant domain references + personas/{id}/accounts.md rolemodel section |
| Content generation | persona.md core/decide/energy/blind + nuance.md voice/platform + selected trend input |
| QA simulation | generated drafts + accounts.md rolemodel/virtual + platform pattern constraints |
| Publishing and tracking | final selected draft + account mapping + latest performance records |
| System or architecture tasks | .ref/PLAN_FOR_POLYSONA.md and this file |
Data Conventions
- Store operational data in Markdown files only.
- Git is the database and history ledger.
- Use PLOON table format:
[table#N](col1,col2,...)then pipe-delimited rows. - Keep persona datasets under
personas/{id}/(persona.md,nuance.md,accounts.md). - Prefer compact, parseable tables over prose for machine-readable persona data.
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 · 52 lines · 623 tokens per session scan A 8d5e349b92f2
polysona CLAUDE.md is an instructions file published in the GitHub repository LilMGenius/polysona (161 stars, last pushed 15d ago), licensed MIT. It adds 623 tokens to every session, about $0.0031 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
bl1nk-agents-manager GEMINI.md
Instructions for bl1nk-bot/bl1nk-agents-manager: For additional context about technologies to be used, project structure, shell commands, and other important information, read the current plan.
intelligent-terminal rust.instructions.md
Concise Rust coding conventions for this repository.
InvestSkill GEMINI.md
Gemini CLI instructions for yennanliu/InvestSkill, covering investskill — gemini cli setup & usage guide, installation & setup, quick start, navigate to the investskill directory and start gemini cli (loads gemini.md automatically).
she-love-me CLAUDE.md
Instructions for 863401402/she-love-me, covering claude.md and 唯一工作流.
slivingdoc AGENTS.md
AGENTS.md instructions for baalimago/slivingdoc, covering agents.md — slivingdoc, architecture, package map, event flow and startup wiring (main.go).
dshcode CLAUDE.md
Claude Code instructions for whitelonng/dshcode, a project described as: Community desktop companion for DeepSeek Harness — one-click Electron app for macOS and Windows.