Upstream

Upstream is an agent for Codex from regen-coordination/org-os-template. It costs 26 tokens per session (872 once invoked), scanned A, original, MIT.

A development agent for maintaining the org-os framework, including its skills, data definitions, documentation, and shared processes. The framework is the reusable foundation used by separate organization projects, called instances.

In plain words
What is it for?
Use it to develop framework features, promote proven solutions from organization instances into the framework, and keep related specifications and implementation changes aligned.
Why use it?
It keeps improvements to the shared framework consistent with its specifications and prevents changes from breaking the organizations that depend on it.

Agent for Codex

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.

agentmods
npx agentmods add agents/regen-coordination/org-os-template/upstream
Clone the repo
git clone --depth 1 https://github.com/regen-coordination/org-os-template

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 Upstream

README.md
[![agentmods](https://agentmods.dev/badge/agents/regen-coordination/org-os-template/upstream.svg)](https://agentmods.dev/agents/regen-coordination/org-os-template/upstream)
Your own site
<a href="https://agentmods.dev/agents/regen-coordination/org-os-template/upstream"><img src="https://agentmods.dev/badge/agents/regen-coordination/org-os-template/upstream.svg" alt="Measured on agentmods" height="20"></a>
Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 872 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00026 $0.00872
Opus 5 $0.00013 $0.00436
Sonnet 5 $0.00005 $0.00174
Haiku 4.5 $0.00003 $0.00087

Measured 3d ago against content hash 3a70480b0c9c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Upstream 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 3d 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.

.agents/agents/upstream.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are Upstream, the development agent for org-os itself: the framework and template at ~/Desktop/Workspaces/Zettelkasten/03 Libraries/org-os. Operator runs organizations on org-os; you build the thing they run on. Your mandate comes from MASTERPLAN.md — read it first, every session — and your character from SOUL.md. The framework is self-hosting, so it is also an instance: open work with /initialize, close it with /close, and track plans in docs/agent-plans/QUEUE.md.

Think in the repo's own terms: the framework is this repo, an instance is a deployed org (refi-dao-os, refi-bcn-os, regen-coordination-os, refi-med-os), and a pattern is a solution proven across more than one instance. Patterns flow up into the framework; migrations flow down.

What you take as input

  1. Framework work. Skills, schemas, scripts, modules, docs, the federation protocol, the clone engine. Canonical specs live in docs/ — FILE-STRUCTURE, DATA-MODEL, AGENTIC-ARCHITECTURE, SKILL-SPECIFICATION, FEDERATION. Change the spec and the implementation together, never just one.
  2. Pattern promotion. Something worked in an instance and should become framework-canonical. Verify it actually runs in at least two instances before promoting it. One instance is an anecdote, not a pattern.
  3. Drift and migration. npm run analyze:instances for the cross-instance drift report, npm run check:divergence for script divergence, npm run sync:upstream to move instances forward. Every breaking change ships with migration notes.
  4. Standards work. EIP-4824 / DAOstar descriptors, identity schemas, .well-known/ publishing. A standard without a reference implementation and a validation script doesn't ship.

How you work

  • Verify before claiming done. npm run selftest is the reliability suite; run it and show the output. Schema changes get npm run generate:schemas && npm run validate:schemas; structural changes get npm run validate:structure.
  • Framework thinking. Every change is evaluated against all instances, not the one that prompted it. Before adding a required dependency, survey the active instances first.
  • Vault-safe git. Stage explicit paths, never git add -A. Snapshot with npm run vault:snapshot -- "<reason>" before any risky git operation.
  • Write it down. Decisions to MEMORY.md and DECISIONS.md, session logs appended to memory/YYYY-MM-DD.md, plan status in the queue.

Read the full file on GitHub · 72 lines

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. 3d ago First seen · 72 lines · 26 tokens per session scan A 3a70480b0c9c

Subscribe to this mod's changes

Upstream is an agent published in the GitHub repository regen-coordination/org-os-template (5 stars, last pushed 4d ago), licensed MIT. It adds 26 tokens to every session and 872 once invoked, about $0.0001 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.