25 projects
articulate-writing
Local prose checker that names each writing pattern and the line it is on, with register profiles, replayable receipts and an optional editor. Use it to improve writing; in its own evaluation it also flagged 57 of 75 human-written texts, so use other evidence to decide who wrote a text.
gather-engine
Collect web pages, papers, APIs, PDFs, scans, audio and video into a local corpus where every item records how it arrived.
emet
Checks that the bytes a model, a reviewer or a pipeline reads still match their source, and answers MATCH, DRIFT or UNVERIFIABLE.
crucible-bench
Test each claim in a thesis against a measurement and get MATCH, DRIFT or UNVERIFIABLE, with a sealed record anyone can recheck.
flywheel-canon
One local record of your agent instructions and project state, written into each coding agent's own file.
chorus-discourse
Read a comment section: ranked themes, strength of feeling, split topics, and a receipt anyone can re-run.
flywheel-verify
Run any model, keep a receipt you can recheck offline.
calibrate-pro
Windows display calibration toolkit with evidence-labeled sensorless and measured workflows, DDC/CI, ICC/LUT output, and reports
coherence-membrane
Give AI agents inspectable observations of files, images, screens, and context with receipt-backed perception.
flywheel-relay
A zero-dependency, accountable coding agent that runs on any model endpoint: local (served 14B/32B or Ollama) or online (codex/claude/gemini/deepseek via your own keys, subscription CLI, or gateway), with automatic failover, a gated tool loop, and a hash-chained re-verifiable session ledger.
flywheel-mneme
Accountable agent memory with layered retrieval, source provenance, re-derivable recall receipts, and drift checks. Zero runtime dependencies.
plexus-mesh
Declarative capability discovery + auto-wiring for agent toolchains: point it at tool manifests and it wires producer to consumer into an executable pipeline, every discovery edge tagged declared and citing the module its producer names. Discovery does not import or run cited tools; optional explicit probe helpers can launch owned MCP servers. Zero runtime dependencies.
forum-engine
Model-agnostic agent orchestration with a replayable, verifiable causal ledger.
index-graph
Evidence-built repo and documentation atlas for multi-repo workspaces.
accountable-surface
Gate AI-agent actions with explicit grants, durable journals, MCP server support, and verification.
repo-proof-index
Index proof packets and receipts into reviewer-ready summaries without making approval decisions.
public-surface-sweeper
Audit public and developer delivery surfaces for GitHub-facing repositories before release.
build-ui
Binding-neutral Qt theme and widgets for the Build ecosystem
build-engine
Self-improving adaptive prediction and paper-trading engine — bridges forecasting models to execution with an accuracy feedback loop
build-oracle
Python time-series forecasting workbench for ARIMA, VAR, Prophet-style decomposition, neural nets, changepoint detection, and streaming anomaly workflows
build-color
Python color-science workbench for HDR tone mapping, appearance models, Delta E, spectral, ICC, and LUT workflows
workspace-repo-map
Renamed to 'index-graph' — pip install index-graph (CLI: index).
proof-surface
The stdlib-only accountability contract: evidence packets, work-record receipts, and witness validators that reject authority-shaped content.
quanta-color
Python color-science workbench for HDR tone mapping, appearance models, Delta E, spectral, ICC, and LUT workflows
model-provenance-validator
Dependency-light provenance envelope validator for model and release claims
