feat(workspace-probe): CLI (readout/probe/grade-mutation) + runnable demo (0.1.1)#136
Merged
Conversation
…demo (0.1.1)
Ships the audit-tooling layer for the interpretability stack (user-directed polish):
- `workspace-probe` bin — a JSON-in/JSON-out CLI over @metaharness/workspace-lens + this package, for
audit workflows that don't want to write TS (composes with jq/CI):
diag <lens.json> — fitted-lens metadata
readout <lens.json> <activations.json> [--top-k N] — workspace tokens per activation
probe <receipts.json> [--drift-threshold F] — workspaceProbeScore over a receipt set
grade-mutation <baseline.json> <mutant.json> — keep/veto verdict
Exit codes 0 ok / 2 usage / 1 runtime; `runCli(argv)` is exported (pure w.r.t. process.exit) for tests.
- examples/demo.mjs — a $0, no-model end-to-end walk-through (synthetic lens → readout → receipts →
probe score → mutation veto). Runs deterministically, no deps.
6 CLI tests (diag/readout/probe/grade-mutation + usage/help/unknown + runtime-error paths) via a stdout
spy; full workspace-probe suite 15/15; tsc clean; healthcheck version green (0.1.1, independently-versioned).
README gains CLI + demo sections. bin added to `files`.
Co-Authored-By: claude-flow <ruv@ruv.net>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What / why
The audit-tooling polish for the interpretability stack (user-directed): a shell surface + a runnable demo so the shipped libraries are usable from the command line and easy to try.
workspace-probebin — JSON-in / JSON-out CLI over@metaharness/workspace-lens+ this package (composes withjq/CI):0ok /2usage /1runtime.runCli(argv)is exported (pure w.r.t.process.exit) so it's unit-testable.examples/demo.mjs— a$0, no-model end-to-end walk-through: synthetic lens → readout → receipts → probe score → mutation veto. Deterministic, no deps (node packages/workspace-probe/examples/demo.mjs).Tests / verify
tscclean; healthcheck version green (0.1.1, independently-versioned).diag/readoutprint correct JSON; the demo vetoes a workspace-degrading mutation (critical rate 0→0.5, clean fraction 1→0.5).Publishes as
@metaharness/workspace-probe@0.1.1on the next tag.🤖 Generated with claude-flow