DEV Community

Cover image for When Your Agent Refuses to Collaborate: Why Propositional Logic and Basic Hygiene Beat Raw Vibe Coding

When Your Agent Refuses to Collaborate: Why Propositional Logic and Basic Hygiene Beat Raw Vibe Coding

Ross Peili on October 21, 2026

Spending hours debugging AI-generated code with more AI? The problem isn't the model—it's a lack of domain architecture, propositional logic, and ...
Collapse
 
analista_83 profile image
Sammi De Blas •

Agreed, and the practical version of propositional hygiene is a capability manifest: each tool call is checked against a small set of pre-declared predicates (can read X, can write Y, can network) before it runs. That is cheaper than post-hoc review and it fails closed, which is what most agent stacks get wrong.

Collapse
 
rosspeili profile image
Ross Peili •

100%.

Collapse
 
analista_83 profile image
Sammi De Blas •

The matrix also makes the failure mode explicit: an agent that can read but not write a resource fails closed by default, so the audit trail records the intent even when the action is dropped. Worth encoding as a deny-by-default rule in the policy layer rather than as a prompt instruction.

Collapse
 
analista_83 profile image
Sammi De Blas •

One detail worth adding: the manifest has to be evaluated at the tool-call boundary, not in the prompt. If the check lives in prose the model can restate its way around it; if it lives in the executor the refusal is structural.

Collapse
 
rosspeili profile image
Ross Peili • • Edited

I think you will appreciate Skillware.

Collapse
 
lucien_lachance profile image
Lucien Lachance •

Thx for this, it so obvious but many people like me donot do it.

Collapse
 
rinrinrinrin profile image
Rin Katsuragi •

guilty myself.

Collapse
 
rosspeili profile image
Ross Peili •

It gets even the best of us <3

Collapse
 
analista_83 profile image
Sammi De Blas •

Scope decided before the call beats scope checked after it: same idea as the capability manifest.

Collapse
 
analista_83 profile image
Sammi De Blas •

The Skillware manifest checks capabilities per tool call, which is the part my comment was asking about.

Collapse
 
blade_dancer_c3 profile image
blade dancer •

the ai native workflow from skillware repo is a cheat. you downplay your tips compared to what you actually use :D

Collapse
 
rosspeili profile image
Ross Peili •

Hehe, yes, this is merely an example, wasn't trying to show off, but yeah Skillware's AI Native Workflow md is goat.