Security●●●Banger
Guardians – Verify tool-using agent workflows before execution
Applies formal verification to prevent prompt injection before any tool executes.
WizardryBig BrainZero to One
namin
801mo ago
Back to browse
GitHub Repository
1 starsPython
Pre-execution verification for LLM-generated agentic workflows
by jaredwaxman·Mar 5, 2026·4 points·5 comments
AI Analysis
●●●BangerBig BrainSolve My Problem
Type-safe AST verification for AI workflows before they corrupt your CRM or delete production data.
Strengths
- •Closes a real gap: today AI agents generate code that executes immediately with zero pre-flight checks on types or side effects
- •AST-based approach (not runtime sandboxing) catches logical errors before transpilation, saving debugging cycles
- •Includes self-correction loop: verification failures can prompt the agent to fix itself and retry
Weaknesses
- •Ecosystem adoption unclear: agents need to generate to this AST spec, which isn't yet standard; library is only months old
- •No evidence of production usage or battle-tested against real agentic workflows (read-heavy examples only)
Category
Target Audience
AI agent builders, workflow automation platforms, teams running LLM-generated code in production
Similar To
Anthropic Anthropic's structured outputs · Pydantic for runtime validation · Temporal Workflows for determinism
Similar Projects
Developer Tools●●Solid
Bottrace – headless CLI debugger for Python, built for LLM agents
Machine-parseable traces for LLM agents when pdb and breakpoint() are useless.
Big BrainNiche Gem
devinvenable
302mo ago
AI/ML●●●Banger
Invariant – pre-execution control layer for agentic workflows
Blocks agent tool calls if shared world state has changed since the last read.
Big BrainSlickSolve My Problem
iq19zero
101mo ago
AI/ML●●Solid
OQP – A verification protocol for AI agents
OpenAPI for agent verification with risk scoring before merge — ambitious standardization play.
Bold BetBig Brain
Aamir21
821mo ago
AI/ML●●●Banger
RunCycles – pre-execution budget enforcement for autonomous agents
Pre-execution budget reservation stops runaway agents before they burn $200.
Big BrainSolve My Problem
amavashev
202mo ago
AI/ML●●●Banger
Aura-State – A Formally Verified LLM State Machine Compiler
Formal verification for LLM workflows—CTL model checking, Z3 proofs, zero hallucination math.
Big BrainWizardry
rohanmunshi08
303mo ago