Argument status

Flag: --status (or --compute).

Where derived confidence asks “how strong?”, argument status asks a sharper question: “does this claim survive every attack?” The answer is one of three labels.

This is the grounded extension from Dung’s argumentation framework (1995).

The attack graph

Only two relations count as attacks:

• opposes — rebuts a node.
• undercuts — defeats an inference.

There’s no separate “defends” relation, because defense falls out for free: defending X means attacking X’s attacker. If risk is attacked by port-strike, and guard opposes port-strike, then port-strike goes out and risk is reinstated to in — automatically.

The labelling

Computed to the least fixpoint:

1. Start every contested node undecided.
2. Repeatedly:
   • label a node in if all of its attackers are already out (a node with no attackers is in immediately);
   • label a node out if any attacker is in.
3. Stop when nothing changes.

The result is unique and deterministic. Nodes caught in an unbroken mutual attack stay undecided.

Worked example

From the bundled pour-the-slab.thml:

link truck-is-booked supports conditions-are-fine
link overnight-freeze opposes conditions-are-fine

overnight-freeze has no attackers → in. It opposes conditions-are-fine, and that attacker is in, so conditions-are-fine → out. (The supports edge doesn’t enter the status calculation — support isn’t an attack; it feeds derived confidence instead.)

That out is exactly what the conflict report compares against the engineer’s authored 0.88.

Where it appears

argument_status is set on every focus and link that takes part in the attack graph:

{ "type": "focus", "id": "conditions-are-fine", "argument_status": "out" }
{ "type": "focus", "id": "overnight-freeze",     "argument_status": "in"  }

It reads as a node colour/badge in the playground’s Argument lens.