ArchitectureThe harness, layer by layer

How the harness runs a mission.

The loop, the specialists, the surfaces they run on, the rules, and what gets written back.

Skyflo provides AI agent orchestration for software engineering through one local mission. The approved plan, specialist coding agents, execution surfaces, review, and memory stay attached to it.

harness / cutawayexploded view

Surfaces. Desktop, browser, terminal; all on one mission.

SF-05 · one mission, several surfaces

Interactive cutaway · hover a layer to inspect

Six states, persisted to disk.

A mission moves from discovery to planning, approval, execution, review, and accepted memory. Its state persists locally between sessions.

state
persisted beyond the session
gate
approval is a state, not a prompt
review
separate mutation-disabled profile
Follow one multi-repo mission through all six states
delegation packet · w1/4mission rl-311
to
repo agent 1 of 4
model slot
“frontier” · routed by you
repository
api-gateway · branch mission/rl-311
objective slice
enforce per-tenant limits at the edge
pinned contract
rate-limit headers v1 · read-only

tools granted

read repositoryedit isolated worktreerun focused checks

out of bounds: unassigned repositories · unrelated credentials · reviewer mutation

exit: bounded work complete · ready for review

Everything outside the packet is out of scope for that agent.

What a specialist actually receives.

The orchestrator splits the approved plan into packets. Each one names a repository, the slice of the objective it covers, the interface contract pinned for the whole mission, the tools that role may call, and the model slot it runs on.

isolation
one boundary per agent
contract
pinned before any code changes
reviewer
independent · writes no code

What the harness can read and write.

Six surfaces, all attached to the same objective. Opening a browser or a shell for a mission does not start a second, unrelated session, and what each role may do on each surface is explicit.

Repositories

local git

readsregistered checkouts · files · history

writesisolated specialist worktrees

local

Browser

mission surface

readsvisible and interactive page state

writesclick · type · navigate when allowed

available

Terminal

local process

readsstdout · stderr · exit state

writescommands within role capability

available

Monitors

mission schedule

readsrecurring check results

writesstatus attached to the mission

available

Automations

local harness

readstask and prior execution context

writesscheduled engineering work

available

Memory

local service

readsscoped · cited · relevant records

writesreview · accept · dismiss · forget

reviewed
one mission · multiple local surfaces · role capability stays explicit

Four rules every mission runs under.

These are properties of the loop rather than settings in a menu, so there is no configuration in which a mission starts with write access or implements before you approve. Each rule leaves a trace you can replay.

default
read-only until the plan says otherwise
approvals
consequence-triggered · yours to widen
audit
every action attributable
reviewer proof
agent-profiles.test.ts
control plane / policy4 standing rules

Start read-only

Every mission opens in discovery. Writes are explicit, later states.

r.01

Gate on consequence

Plan mode blocks until you approve the implementation boundary.

r.02

Execute locally

Agents run on your linked Mac with your repositories and local credentials.

r.03

Record everything

Actions, prompts, and approvals land in the mission log, replayable.

r.04

audit / rl-311

planplan proposed · implementation blocked

approveplan approved · boundary pinned

reviewread-only reviewer report attached

knowledge / write-backon mission complete

+outcomerate limits enforced at api-gateway · limits.yaml per plan tier

+constraint429 responses must carry Retry-After · v1 clients surface it

~ownershipplatform-infra owns the redis cluster (updated)

proposed on close · each record keeps its source

What the mission leaves behind.

As a mission closes it proposes records: outcomes, constraints, ownership facts. Each one keeps a link to the work that produced it, and none of them is kept until you accept it.

Questions

How the boundaries work.

What persists, when implementation starts, and what each role is allowed to do. To see the same parts in one example, follow an illustrative mission.

01How does Skyflo orchestrate AI coding agents?+

An approved mission plan is divided into bounded work for specialists. Each specialist receives only the repositories and capabilities its part requires, and a separate read-only reviewer checks the combined result.

02What persists?+

The mission objective and state survive the session, so work can continue without becoming a new chat-shaped project.

03When does implementation start?+

Plan mode blocks until you approve the plan. The approved boundary then guides specialist work.

04What can specialists use?+

The current harness exposes code, browser, terminal, monitors, and automations according to each role's capabilities.

05What does independent review mean?+

The reviewer receives only file.read, file.search, and agent.report, receives read-only capability operations, and has mutationAllowed set to false.

Start with one objective.

Install the local harness, use a supported signed-in harness or provider key, and approve the plan before implementation begins.

Local execution, with no markup on BYOK or provider-direct usage.