Commits

Commits on Sep 5, 2026

Merge pull request #6 from METR/feat/citing-scanner
pipmc
authored
06e6f92
refactor: parametrize transcript fixtures directly rather than via flags
pipmc
committed
f535de0
fix: annotate parametrized transcript fixtures for list invariance
pipmc
committed
a68e91b
test: cover both directions of independent reference-kind population
pipmc
committed
3f1ce24
docs: trim citing_scanner docstring to its summary line
pipmc
committed
82be706
fix: address citing_scanner review findings
pipmc
committed
7302b9d

Commits on Sep 4, 2026

feat: add citing_scanner emitting message and event references
pipmc
committed
12b3b7b

Commits on Sep 2, 2026

Merge pull request #5 from METR/feat/hardcoded-429-simulation
rasmusfaber
authored
1842b14
Drop the eval-based rate-limit tests

rasmusfaber
and
claude
committed
0d8561a
Simulate SDK-swallowed retries, and test #5138 properly

rasmusfaber
and
claude
committed
97cf8ac
Record the ModelCall for refused requests

rasmusfaber
and
claude
committed
aca126d
Simulate 429s so adaptive concurrency can be tested

rasmusfaber
and
claude
committed
350f810

Commits on Aug 21, 2026

feat: let configurable_sandbox request a runtime class (#4)

revmischa
and
claude
authored
b65e82e

Commits on Aug 10, 2026

Commits on Jul 20, 2026

Merge pull request #1 from METR/pip/add-failing-scanner
pipmc
authored
f15f30f
Add failing_scanner for deterministic scan-error testing

pipmc
and
claude
committed
adb1e76
Initial commit
pipmc
committed
e3606b4