Project Telos is looking for field testers for index: rerunnable workspace maps and evidence-built repo/documentation atlases.
The question this repo should help answer:
Can a multi-repo workspace map be rerun later and show whether the architecture claim still matches the filesystem?
Useful test lanes:
- DevOps/DevSecOps: map service repos, scripts, CI files, and docs, then check whether handoff notes match actual state.
- Supply-chain/security: verify whether a repo claim is backed by files, tests, manifests, and dependency surfaces.
- ML/backend infrastructure: map model-serving or data-pipeline repos so another reviewer can inspect the boundary.
- Agent workflows: give an agent a repo map and see whether it stops inventing files or dependencies.
What I am looking for:
- Verification against real public repos or synthetic workspaces.
- Cases where the map misses an important surface.
- False confidence: places that should be
DRIFT or UNVERIFIABLE, not treated as a match.
- Better output shapes for humans reviewing agent work.
- Early traction and pointers to modest grassroots research funding for checkable-state tooling.
Project links:
Stage label: public, solo, early, pre-revenue, not independently audited. Please start with public repos or synthetic test workspaces.
Project Telos is looking for field testers for
index: rerunnable workspace maps and evidence-built repo/documentation atlases.The question this repo should help answer:
Useful test lanes:
What I am looking for:
DRIFTorUNVERIFIABLE, not treated as a match.Project links:
Stage label: public, solo, early, pre-revenue, not independently audited. Please start with public repos or synthetic test workspaces.