CalvinBackup/NeuroSploit
1
0
Fork 0
mirror of https://github.com/CyberSecurityUP/NeuroSploit.git synced 2026-09-29 12:31:43 +02:00
Code Issues Packages Projects Releases Wiki Activity

feat(typesafe): System One calibrated adjudication (RLCD tier)

Browse Source 
Integrates TypeSafe's System One model (Jev) as an optional, calibrated
adjudicator — the RLCD (Reinforcement Learning for Calibrated Decisions) tier:
typed judgments with probabilities where the harness needs a number, not prose.

- typesafe.rs: HTTP client for POST /v1/systemone (Bearer TYPESAFE_API_KEY,
  model jev-latest), with Choice/Noul/Score primitives, retry on 429/529, and
  parsed answers exposing the probability distribution + confidence.
  adjudicate() asks a Choice {confirmed/needs-review/rejected} plus an
  impact-demonstrated Noul over a finding; calibrated_confidence() folds
  demonstrated impact into the number, wants_review() gates a split distribution.
- pipeline: an optional pass (runs when TYPESAFE_API_KEY is set, off with
  NEUROSPLOIT_TYPESAFE=off) adjudicates each finding over its EVIDENCE — never
  its narrative — refining confidence and the needs-review boundary. Additive:
  a deterministic validator still rules; TypeSafe can only lower confidence or
  flag for review, never resurrect a rejected claim. Audited per finding.
- env.example + README document it; the web console inherits the key via env.

Where the model stack maps in NeuroSploit: LM/BERT ≈ the deterministic
validators (no model), RLHF chat ≈ the exploit/recon agents, RLVR reasoning ≈
the DeepReasoning budget tier, RLCD ≈ this calibrated adjudication.

373 tests (+5).

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
CyberSecurityUP
2026-09-19 17:56:34 -03:00
co-authored by Claude Opus 5
parent 2e95556df5
commit c3de51d508
5 changed files with 415 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.env.example
+7
View File
@@ -68,3 +68,10 @@ NOUS_API_KEY=
# model through it as `--model litellm:<model>`.
#LITELLM_BASE_URL=http://localhost:4000/v1
LITELLM_API_KEY=
# TypeSafe System One (RLCD — Reinforcement Learning for Calibrated Decisions).
# When set, NeuroSploit adjudicates each finding with a calibrated verdict
# (confirmed/needs-review/rejected) + an "impact demonstrated" judgment over the
# EVIDENCE, refining confidence and the needs-review boundary. Additive — never
# overrules a deterministic validator. Disable with NEUROSPLOIT_TYPESAFE=off.
TYPESAFE_API_KEY=