turritopsis-persistent-agent-runtime

turritopsis-persistent-agent-runtime is a skill for Codex from anhe2021212-spec/Turritopsis. It costs 82 tokens per session (275 once invoked), scanned A, original, MIT.

A guide for designing and repairing Turritopsis, a long-running agent system that keeps working across sessions. It covers the separate parts that carry messages, run tools, schedule work, store queues, and maintain continuity.

In plain words
What is it for?
Use it to review runtime architecture, define layered health checks, investigate relays and channels, and plan recovery for a persistent agent.
Why use it?
It prevents operators from treating a restart, a healthy process, and a working agent as the same thing. It also helps avoid losing the agent's identity or context when sessions change.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: $skill-name invocation.

Good fit Use it to review runtime architecture, define layered health checks, investigate relays and channels, and plan recovery for a persistent agent.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/anhe2021212-spec/turritopsis/turritopsis-persistent-agent-runtime
Install

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.

Any agent
npx skills add anhe2021212-spec/Turritopsis --skill turritopsis-persistent-agent-runtime
Clone the repo
git clone --depth 1 https://github.com/anhe2021212-spec/Turritopsis

Made for: Codex.

Wrote 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.

agentmods badge for turritopsis-persistent-agent-runtime

README.md
[![agentmods](https://agentmods.dev/badge/skills/anhe2021212-spec/turritopsis/turritopsis-persistent-agent-runtime/github.svg)](https://agentmods.dev/skills/anhe2021212-spec/turritopsis/turritopsis-persistent-agent-runtime)
Your own site
<a href="https://agentmods.dev/skills/anhe2021212-spec/turritopsis/turritopsis-persistent-agent-runtime"><img src="https://agentmods.dev/badge/skills/anhe2021212-spec/turritopsis/turritopsis-persistent-agent-runtime/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for turritopsis-persistent-agent-runtime

Your own site · 80×15
<a href="https://agentmods.dev/skills/anhe2021212-spec/turritopsis/turritopsis-persistent-agent-runtime"><img src="https://agentmods.dev/badge/skills/anhe2021212-spec/turritopsis/turritopsis-persistent-agent-runtime.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 82 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 275 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00082 $0.00275
Opus 5 $0.00041 $0.00138
Sonnet 5 $0.00016 $0.00055
Haiku 4.5 $0.00008 $0.00028

Measured 12d ago against content hash d0215dbade60, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

turritopsis-persistent-agent-runtime 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 12d 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.

skills/turritopsis-persistent-agent-runtime/SKILL.md · 18 lines

What it actually says

Turritopsis Persistent Agent Runtime

Map the Agent's operating body without confusing it with personality or memory.

  1. Use $turritopsis-onboarding for local scanning and validated skeleton application.
  2. Read stage-suite.md.
  3. Identify the live runtime authority before trusting configuration, documentation, a local clone, or a review mirror.
  4. Separate model session, process, relay, channel, scheduler, tool layer, queue, and durable state by responsibility.
  5. Protect identity or context continuity with an explicit boundary; do not hide session replacement inside a generic restart procedure.
  6. Define health in layers: process, transport, channel, tool path, and end-to-end function.
  7. Record degraded or offline state as unknown where evidence is unavailable; do not invent a failure.
  8. Use fixture.json as a fictional quality example only.
Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 12d ago First seen · 18 lines · 82 tokens per session scan A d0215dbade60

Subscribe to this mod's changes

turritopsis-persistent-agent-runtime is a skill published in the GitHub repository anhe2021212-spec/Turritopsis (13 stars, last pushed 17d ago), licensed MIT. It adds 82 tokens to every session and 275 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

opensrc

Fetch dependency source code to give AI agents deeper implementation context. Use when the agent needs to understand how a library works internally, read source code for a package, fetch implementation details for a dependency, or explore how an npm/PyPI/crates.io package is built. Triggers include "fetch source for"…

vercel-labs/opensrc · 103 tokens

architecture

Modular design, interface stability, dependency management, AI-code health. Triggers: architecture, design, structure, modules, dependencies, coupling, system design.

ariaxhan/kernel-claude · 34 tokens

using-goodmemory

Recall and persist durable project context with GoodMemory in Kimi Code.

hjqcan/GoodMemory · 18 tokens

outcome-run

Run one action in an outcome-loop folder. Use when the user wants to execute the outcome loop, start with Superdense reward maintenance, choose one action on a lever, create runs/ /work.md and learnings.md, or use prior outcomes to improve a real-world metric.

Nimrobo/superdense · 62 tokens

outcome-setup

Set up or repair one outcome-loop folder for improving a real-world outcome. Use when the user wants to define goal.md, run.md, analytics/instrumentation, levers, target repos/accounts, or the folder contract for an outcome loop.

Nimrobo/superdense · 54 tokens

outcome-update

Improve an existing outcome loop playbook from prior runs and Superdense reward evidence. Use when the user wants observer-mode analysis, macro workflow improvements, run.md updates, lever-map refinement, diagnostics improvements, or efficiency fixes for future outcome runs.

Nimrobo/superdense · 52 tokens