Skip to content
View taipei49314's full-sized avatar
🧭
Preparing Frontier Atlas for public semantic-audit testing.
🧭
Preparing Frontier Atlas for public semantic-audit testing.

Block or report taipei49314

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
taipei49314/README.md

Nelson — evidence-first agent systems

I build local-first, deterministic, auditable tools for AI agents and developer workflows.

我在台灣打造本機優先、可重播、可稽核的 AI agent 與開發工具。

Models may propose. Verifiers decide. Missing evidence stays UNKNOWN / INCOMPLETE.

Current focus: Frontier Atlas

Frontier Atlas is a private product and research line for auditing whether an agent's claim is actually supported by its cited evidence.

frontier-atlas-open-tests is its public, offline test surface:

  • bounded claim–citation audit schemas;
  • deterministic protocol and identity verification;
  • blind, non-gold calibration packets;
  • commit → reveal → adjudication workflows;
  • reproducible public issue intake.

Frontier Atlas offline tests

The current release stage is P4.5-T: external tester preparation. Promotion still requires a second, independent natural person; that rule is not replaced by two models, two sessions, or two signing keys.

The planned path is:

external qualificationdouble-blind pilot + real-source calibrationprivate judge/token optimizationdevelopment benchmarksealed holdoutblack-box closed betarelease candidateexternal evidence reviewproduction

Public visibility grants neither a passing result nor access to private product core, hidden labels, qualification keys, human records, or sealed holdout authority. A repository is open source only when its own license explicitly says so.

Public tools

Project Purpose
frontier-atlas-open-tests Public offline qualification and semantic-audit test surface
greenwash Detects when an agent makes CI green by weakening verification
nullbench Pre-registers decisions and scores them against chance
trust-meter Deterministic, evidence-backed trust scoring
unasked Evidence-gated repository investigation; non-certifying
branchback Replays belief-at-the-time against knowledge-now
aurora Evidence-led industry discovery without an LLM at runtime
receiptradar Local receipt-to-ledger CLI with no cloud account
universe-explorer Epistemically honest science knowledge system

Engineering rules

  1. Evidence over confidence. Claims point to exact citations, tests, diffs, logs, artifacts, or an explicit insufficient-data result.
  2. Fail closed. Missing observation is not a pass.
  3. Deterministic first. The same frozen evidence should produce the same mechanical result.
  4. Authority stays separate. A model can propose; it cannot silently grant review, qualification, or release authority to itself.
  5. Keep negative results. Blocked gates and counterexamples remain visible rather than being rewritten as success.
  6. Measure cost with real denominators. Speed, token, and accuracy claims wait for reproducible benchmark data.

Stack

Python · Rust · Go · TypeScript · FastAPI · React · SQLite · offline-first CLI workflows

Public portfolio reconciled 2026-08-29. Current status: preparing Frontier Atlas for public semantic-audit testing.

Pinned Loading

  1. greenwash greenwash Public

    Catch the agent that made CI green by weakening the tests. Deterministic, zero-LLM, zero-network, sub-second — reads the diff, not the code state. Public bypass list included.

    Python

  2. universe-explorer universe-explorer Public

    Honestly separating what we know from what we don't — an epistemically honest science knowledge system. 誠實區分已知與未知的科學知識系統。

    Python 1

  3. frontier-atlas-open-tests frontier-atlas-open-tests Public

    Public offline qualification and semantic-audit test surface for Frontier Atlas

    Python

  4. nullbench nullbench Public

    Pre-register decisions. Score them against chance. Never backfill.

    Python

  5. trust-meter trust-meter Public

    Measure before you trust. Deterministic, local-first, evidence-backed trust scoring.

    Python

  6. unasked unasked Public

    Evidence-gated repository investigation research harness (non-certifying alpha)

    Python