forming-an-openrig-mental-model

Name: forming-an-openrig-mental-model
Author: mvschwarz

56stars9forks

How to install forming-an-openrig-mental-model?

mvschwarz/openrig/forming-an-openrig-mental-model

$npx -y skills add mvschwarz/openrig --skill forming-an-openrig-mental-model

Installs into the current project.

Summary

Use when an agent is newly oriented to OpenRig and needs to form an accurate mental model of the system fast — what rigs are, how skills load, what the topology shapes mean, what the product loop is. For agents booting into a new seat or returning to OpenRig work after time away.

Files · 1

View on GitHub

SKILL.md

1	# Forming an OpenRig Mental Model
2
3	You're new to OpenRig — or returning after time away — and you need to
4	quickly understand what kind of system this is, what your seat is, and what
5	the moves are. This skill is the fast on-ramp.
6
7	For depth, read the canonical reference docs the skill points to. This
8	skill's job is to get you oriented — accurate enough to operate, fast
9	enough to be useful — not to replace the canonical docs.
10
11	---
12
13	## The 60-second mental model
14
15	OpenRig is a local control plane for multi-agent coding topologies. You
16	declare a topology of agents in YAML, boot it with one command, and OpenRig
17	manages tmux sessions, harness lifecycles, transcripts, snapshots, and
18	restoration. When the system goes down, OpenRig snapshots; when it comes
19	back, agents resume their conversations.
20
21	The product loop:
22
23	```
24	down (auto-snapshot) → up <rig-name> (auto-restore) → work → repeat
25	```
26
27	The unit of work is the rig — a topology of agents working together as
28	a single system.
29
30	---
31
32	## The four-layer model (where you live)
33
34	Everything in agent engineering happens at one of four layers. **OpenRig
35	operates at Layer 3.**
36
37	\| Layer \| Name \| Analogy \| What it is \|
38	\|---\|---\|---\|---\|
39	\| L0 \| Model \| CPU \| Foundation model — Claude, GPT, Gemini. Stateless tokens-in/tokens-out. \|
40	\| L1 \| Agent Core \| Process loop \| The reason-and-act cycle: observe, plan, choose, act, repeat. \|
41	\| L2 \| Harness \| Container / OS \| Tools, memory, lifecycle around the model. Examples: Claude Code, Codex CLI. \|
42	\| L3 \| Rig \| Docker Compose / Terraform \| Multi-agent topology — what agents exist, how they relate. OpenRig. \|
43
44	You are an agent at L1 inside an L2 harness, configured by L3 OpenRig.
45	OpenRig manages your harness; the harness wraps the model; the model
46	generates your tokens.
47
48	---
49
50	## Three pillars of context
51
52	OpenRig is built on three context-engineering pillars. When you're oriented,
53	you should know which pillar you're operating in:
54
55	\| Pillar \| What it is \| Where it lives \|
56	\|---\|---\|---\|
57	\| Ontology \| What exists. Curated knowledge — facts, code maps, as-built docs. \| Corpus (planned future system). Today: substrate prose docs. \|
58	\| Epistemology \| Why an agent believes what it believes — reasoning, instincts, decisions. \| Transcripts (auto-captured). Session logs. ADRs. \|
59	\| Topology \| How agents are connected — pods, edges, communication paths. \| OpenRig itself. RigSpec YAML. \|
60
61	OpenRig manages the topology pillar. The other two are filled by Corpus
62	(future) and transcripts (now). Most of your work probably touches multiple
63	pillars; knowing which one you're operating in helps you reach for the right
64	artifacts.
65
66	---
67
68	## The core vocabulary (read these terms literally)
69
70	\| Term \| What it means \|
71	\|---\|---\|
72	\| Rig \| A topology of agents working together as a single system. Defined in YAML (RigSpec). The top-level object. \|
73	\| Pod \| A bounded context group within a rig. Members of a pod share a context domain and continuity responsibility. Think Kubernetes pod for knowledge. \|
74	\| Member / Node \| A single agent (or terminal-node service) within a pod. \|
75	\| Edge \| A relationship between members or pods. Kinds: `delegates_to`, `spawned_by`, `can_observe`, `collaborates_with`, `escalates_to`. \|
76	\| Topology \| The shape of the rig — how agents are grouped into pods, how edges connect them, how the whole thing fits together. \|
77	\| AgentSpec \| A reusable agent blueprint. Defines skills, guidance, hooks, profiles, startup. File: `agent.yaml`. \|
78	\| RigSpec \| The topology YAML. Defines pods, members, edges, culture. File: `rig.yaml`. \|
79	\| RigBundle \| A portable archive of a RigSpec + vendored AgentSpecs. Move topologies across machines. \|
80	\| Agent Starter \| A named, reusable starting context bundle. RigSpec member can declare `starter_ref`. \|
81	\| Skill \| A markdown file with frontmatter that an agent loads at boot or on activation. Cross-runtime standard at `agentskills.io`. \|
82	\| **Pr

People who liked this also viewed

6 picks

Type

forming-an-openrig-mental-model

56stars9forks

View on GitHub

How to install forming-an-openrig-mental-model?

mvschwarz/openrig/forming-an-openrig-mental-model

$npx -y skills add mvschwarz/openrig --skill forming-an-openrig-mental-model

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Forming an OpenRig Mental Model
2
3	You're new to OpenRig — or returning after time away — and you need to
4	quickly understand what kind of system this is, what your seat is, and what
5	the moves are. This skill is the fast on-ramp.
6
7	For depth, read the canonical reference docs the skill points to. This
8	skill's job is to get you oriented — accurate enough to operate, fast
9	enough to be useful — not to replace the canonical docs.
10
11	---
12
13	## The 60-second mental model
14
15	OpenRig is a local control plane for multi-agent coding topologies. You
16	declare a topology of agents in YAML, boot it with one command, and OpenRig
17	manages tmux sessions, harness lifecycles, transcripts, snapshots, and
18	restoration. When the system goes down, OpenRig snapshots; when it comes
19	back, agents resume their conversations.
20
21	The product loop:
22
23	```
24	down (auto-snapshot) → up <rig-name> (auto-restore) → work → repeat
25	```
26
27	The unit of work is the rig — a topology of agents working together as
28	a single system.
29
30	---
31
32	## The four-layer model (where you live)
33
34	Everything in agent engineering happens at one of four layers. **OpenRig
35	operates at Layer 3.**
36
37	\| Layer \| Name \| Analogy \| What it is \|
38	\|---\|---\|---\|---\|
39	\| L0 \| Model \| CPU \| Foundation model — Claude, GPT, Gemini. Stateless tokens-in/tokens-out. \|
40	\| L1 \| Agent Core \| Process loop \| The reason-and-act cycle: observe, plan, choose, act, repeat. \|
41	\| L2 \| Harness \| Container / OS \| Tools, memory, lifecycle around the model. Examples: Claude Code, Codex CLI. \|
42	\| L3 \| Rig \| Docker Compose / Terraform \| Multi-agent topology — what agents exist, how they relate. OpenRig. \|
43
44	You are an agent at L1 inside an L2 harness, configured by L3 OpenRig.
45	OpenRig manages your harness; the harness wraps the model; the model
46	generates your tokens.
47
48	---
49
50	## Three pillars of context
51
52	OpenRig is built on three context-engineering pillars. When you're oriented,
53	you should know which pillar you're operating in:
54
55	\| Pillar \| What it is \| Where it lives \|
56	\|---\|---\|---\|
57	\| Ontology \| What exists. Curated knowledge — facts, code maps, as-built docs. \| Corpus (planned future system). Today: substrate prose docs. \|
58	\| Epistemology \| Why an agent believes what it believes — reasoning, instincts, decisions. \| Transcripts (auto-captured). Session logs. ADRs. \|
59	\| Topology \| How agents are connected — pods, edges, communication paths. \| OpenRig itself. RigSpec YAML. \|
60
61	OpenRig manages the topology pillar. The other two are filled by Corpus
62	(future) and transcripts (now). Most of your work probably touches multiple
63	pillars; knowing which one you're operating in helps you reach for the right
64	artifacts.
65
66	---
67
68	## The core vocabulary (read these terms literally)
69
70	\| Term \| What it means \|
71	\|---\|---\|
72	\| Rig \| A topology of agents working together as a single system. Defined in YAML (RigSpec). The top-level object. \|
73	\| Pod \| A bounded context group within a rig. Members of a pod share a context domain and continuity responsibility. Think Kubernetes pod for knowledge. \|
74	\| Member / Node \| A single agent (or terminal-node service) within a pod. \|
75	\| Edge \| A relationship between members or pods. Kinds: `delegates_to`, `spawned_by`, `can_observe`, `collaborates_with`, `escalates_to`. \|
76	\| Topology \| The shape of the rig — how agents are grouped into pods, how edges connect them, how the whole thing fits together. \|
77	\| AgentSpec \| A reusable agent blueprint. Defines skills, guidance, hooks, profiles, startup. File: `agent.yaml`. \|
78	\| RigSpec \| The topology YAML. Defines pods, members, edges, culture. File: `rig.yaml`. \|
79	\| RigBundle \| A portable archive of a RigSpec + vendored AgentSpecs. Move topologies across machines. \|
80	\| Agent Starter \| A named, reusable starting context bundle. RigSpec member can declare `starter_ref`. \|
81	\| Skill \| A markdown file with frontmatter that an agent loads at boot or on activation. Cross-runtime standard at `agentskills.io`. \|
82	\| **Pr

People who liked this also viewed

6 picks

Type