arkheionx blind-spots
Rank likely blind-spot candidates: high-impact surfaces with weak review evidence, scored transparently (impact + review-gap + complexity + assumption).
V5 — Blind Spot Intelligence
Arkheionx v5 adds Blind Spot Intelligence: a local/static way to prioritize high-impact surfaces with weak review evidence. V4 helps answer "how does this protocol work?". V5 helps answer "what are we probably missing?". It does not predict vulnerabilities, assign severity, or run anything against a live chain.
Workflow
arkheionx review-map .
arkheionx blind-spots .
arkheionx criticality-map .
arkheionx counterfactuals .
arkheionx research-pack . --out .arkheionx/research-pack Rank likely blind-spot candidates: high-impact surfaces with weak review evidence, scored transparently (impact + review-gap + complexity + assumption).
Map criticality potential — heuristic blast radius if a bug existed — across surfaces. Not severity.
Negate guarding assumptions into testable "what if this is false?" research prompts with local test directions and stop conditions.
Bundle everything into one local, vendor-agnostic research pack for a human reviewer or an AI agent.
What it surfaces
BSP-001 ClaimGate.claimWithProof
Criticality potential very-high (blast radius, not severity)
Review density none (no direct test observed)
Blind spot score 166 (impact 85 + review-gap 30 + complexity 36 + assumption 15)
Suggested counterfactual a proof for one claim authorizes a different claim shape
Do not claim not a finding without independent local proof Blind spot candidates are places to look. Criticality potential is heuristic blast radius, never a severity. Counterfactuals are research prompts, not findings.
No RPC, no live-chain action, no exploit automation, no auto-submit. Artifacts are written locally under .arkheionx/.
Blind Spot Intelligence allocates attention. A human, with independent local proof, makes every security call. ArkheionX does not confirm vulnerabilities.
Boundary
It does not predict bugs, assign severity, or replace an audit. Read the full guide in the
repository's docs/BLIND_SPOT_INTELLIGENCE.md and docs/V5_WORKFLOW.md.