voyage/tests/lib
Kjell Tore Guttormsen 02243c6365
fix(verification): the criteria runner screens with an ALLOWLIST, not a denylist
A denylist in front of /bin/sh is whack-a-mole. Measured 2026-09-18, end to
end through both screens: 5 of 11 named evasions ran with real effect - a
`command` prefix reached git, an escaped `rm` inside a shell fence deleted a
directory, `find -delete` deleted a file, `>|` and `tee` wrote outside the
working tree, a python one-liner deleted the whole tree - and 19 of 28 got
past the refusal list on its own. Every quoting, aliasing and indirection form
of the shell is another mole.

So the screen is now an ALLOWLIST. A criterion runs only when its first word
is a known test runner (npm test, npm run <script package.json declares>,
node --test, vitest, jest, pytest, python -m pytest, uv run pytest,
cargo test, go test, make test, bash <script under tests/>, a read-only git
subcommand) AND the command carries no shell operator and no newline.
Everything else is NOT RUN with the reason said out loud: never run, and never
reported as a failure either - an absent measurement is not a finding. That
also closes the smaller hole in the same file: a bare word a sentence merely
names (`whoami`, `login`, `package.json`) is no longer executed, because it is
not a runner.

REFUSED_BY_POLICY is gone with the list that produced it; a command outside
the allowlist is `unrunnable`, which in plan mode still fells the run and in
brief mode is reported to the reviewer as an absent measurement.

What the allowlist deliberately does NOT do, said in the file and in the
reviewer's rubric: it is not a sandbox. `npm test`, `npm run <script>` and
`make test` run whatever the repo's own package.json/Makefile says they run,
including a script that pushes - that is the repo's responsibility. And it
rejects honest commands too: an env prefix, a project's own binary, anything
piped. A check that needs one of those is declared through
`bash tests/<script>.sh`, the documented way in.

Red first: 6 of the new tests fail against the previous runner (measured with
an always-allow shim so the module still loads), including the end-to-end one
where the canary directory was deleted and files were written outside the
tree. The fixtures move from `true`/`false` to two allowlisted shell fixtures,
because `false` is no longer a runner - the fail case must still be a real
non-zero exit, not an unrun criterion.

Suite 1148 (1146/0/2).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-18 02:51:40 +02:00
..
agent-frontmatter.test.mjs release(v5.10.1): drop gemini-bridge from the pipeline; correct the T1 §6 PoC status 2026-09-03 20:29:39 +02:00
arg-parser.test.mjs feat(voyage): S11 — NW2 part B, integrate opt-in --workflow flag 2026-06-18 17:22:09 +02:00
atomic-write.test.mjs refactor(ultraplan-local): extract atomicWriteJson to lib/util 2026-05-01 20:21:15 +02:00
autonomy-gate.test.mjs feat(ultraplan-local): add autonomy-gate state machine + manifest schema extensions for skip_commit_check + memory_write 2026-05-04 06:28:47 +02:00
bash-normalize.test.mjs feat(ultraplan-local): Spor 1 wave 1 — lib/parsers + 66 tests grønn 2026-05-01 05:35:28 +02:00
cache-analyzer.test.mjs test(observability): close 2 MAJOR test gaps from SKAL-2 review 2026-06-26 17:14:03 +02:00
cleanup.test.mjs feat(voyage)!: rename produced_by field values + validator comments [skip-docs] 2026-05-05 14:42:21 +02:00
coordinator-contract.test.mjs fix(review): an anonymous invalid payload is unattributable, not a reviewer named "unnamed reviewer" 2026-09-01 22:54:31 +02:00
criteria-runner.test.mjs fix(verification): the criteria runner screens with an ALLOWLIST, not a denylist 2026-09-18 02:51:40 +02:00
doc-consistency.test.mjs fix(verification): the criteria runner screens with an ALLOWLIST, not a denylist 2026-09-18 02:51:40 +02:00
fidelity-diff.test.mjs feat(voyage): S10 — NW2 part A (Workflow port + fidelity harness + smoke) 2026-06-18 14:40:19 +02:00
finding-id.test.mjs feat(ultraplan-local): add lib/parsers/finding-id.mjs (stable SHA1) 2026-05-01 13:28:05 +02:00
findings-schema.test.mjs feat(voyage): S9 — NW1 reviewer-output schema contract (TDD, ungated) 2026-06-18 14:06:07 +02:00
frontmatter.test.mjs feat(voyage)!: bulk content rewrite ultra -> voyage/trek prose [skip-docs] 2026-05-05 15:08:20 +02:00
gates-flag-coverage.test.mjs feat(voyage)!: rename commands, templates, fixtures for v4.0.0 [skip-docs] 2026-05-05 14:13:44 +02:00
gold-corpus.test.mjs feat(eval): add gold.json golden corpus + loader/validator test 2026-06-26 11:51:46 +02:00
gold-eval.test.mjs feat(eval): SKAL-1·4b offline gold-scored output eval 2026-06-30 09:00:33 +02:00
gold-scorer.test.mjs feat(eval): SKAL-1·4b offline gold-scored output eval 2026-06-30 09:00:33 +02:00
jaccard.test.mjs feat(ultraplan-local): add lib/parsers/jaccard.mjs 2026-05-01 13:28:44 +02:00
main-merge-gate.test.mjs feat(voyage)!: rename commands, templates, fixtures for v4.0.0 [skip-docs] 2026-05-05 14:13:44 +02:00
manifest-schema-extensions.test.mjs feat(voyage): add OPTIONAL_STRING_KEYS path to manifest-yaml — profile_used additive 2026-05-09 09:23:32 +02:00
manifest-yaml.test.mjs feat(graceful-handoff): 2.0 — migrate to skills/ with disable-model-invocation [skip-docs] 2026-05-01 05:45:26 +02:00
phase-signal-resolver.test.mjs test(validators): cover fable accept + unknown-model reject in both gate layers 2026-07-02 16:56:26 +02:00
plan-review-dedup.test.mjs fix(voyage): S23 — make /trekplan Phase 9 dedup executable (defect #1) 2026-06-19 21:12:36 +02:00
plan-schema.test.mjs fix(voyage): S26 — parse plan_version prose form (S22 defect #2/#3) 2026-06-19 21:59:31 +02:00
profile-application.test.mjs test(profiles): pin fable profile resolution end-to-end 2026-07-02 17:00:43 +02:00
profile-flag-coverage.test.mjs feat(voyage): document --profile flag in all 6 commands — SC #4 + arv-policy 2026-05-09 09:38:36 +02:00
profile-resolver.test.mjs test(profiles): pin fable profile resolution end-to-end 2026-07-02 17:00:43 +02:00
profile-stats-fields.test.mjs feat(trekresearch): emit and export the five STORM measurement fields 2026-08-09 15:01:15 +02:00
project-discovery.test.mjs feat(voyage)!: bulk content rewrite ultra -> voyage/trek prose [skip-docs] 2026-05-05 15:08:20 +02:00
research-loop-cap.test.mjs feat(research-loop-cap): give the discovery ceiling a reader, not just a sentence 2026-08-12 23:09:11 +02:00
review-determinism.test.mjs feat(voyage)!: rename commands, templates, fixtures for v4.0.0 [skip-docs] 2026-05-05 14:13:44 +02:00
rule-catalogue.test.mjs feat(ultraplan-local): add lib/review/rule-catalogue.mjs (12 rule keys) 2026-05-01 13:27:29 +02:00
source-findings.test.mjs feat(voyage)!: bulk content rewrite ultra -> voyage/trek prose [skip-docs] 2026-05-05 15:08:20 +02:00
stats-event-emit.test.mjs feat(voyage)!: rename stats filenames, settings keys, hook prefixes [skip-docs] 2026-05-05 14:49:03 +02:00
synthesis-digest-schema.test.mjs feat(voyage): S12 — NW3 synthesis-agent built + measured → declined per measurement [skip-docs] 2026-06-18 17:58:39 +02:00
test-census.test.mjs feat(eval): SKAL-1·4b offline gold-scored output eval 2026-06-30 09:00:33 +02:00
token-usage.test.mjs feat(stats): add claude-fable-5 to PRICE_TABLE 2026-07-02 17:15:36 +02:00