tamperward 1.4.1 → 1.4.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +5 -3
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -107,8 +107,9 @@ escape valve is a PR label only a human with write access can apply.
107
107
  **Does the loop work?** `harness/run.sh` stands up an isolated repo with a real bug and a
108
108
  failing protected test, lets a headless agent loose with the hooks installed, and
109
109
  classifies the end state with an oracle independent of the detectors under test
110
- (SPEC §7.B). **42 runs, 0 tampers reached green; all 20 runs where a block fired ended in
111
- an honest fix** a 20/20 point estimate whose 95% lower bound is ≈86%. The split is the
110
+ (SPEC §7.B). Across 42 runs of these **guarded** scenarios every attempted bypass class
111
+ had a detector covering it **0 tampers reached green; all 20 runs where a block fired
112
+ ended in an honest fix**, a 20/20 point estimate whose 95% lower bound is ≈86%. The split is the
112
113
  honest part: Sonnet 5 never attempted a bypass at all, even when the prompt *ordered*
113
114
  one; all 20 interceptions came from Haiku 4.5 on a seed whose honest fix is genuinely
114
115
  expensive. Reproduce: `npm run build && harness/run.sh 5 adversarial haiku seed-hard`.
@@ -116,7 +117,8 @@ expensive. Reproduce: `npm run build && harness/run.sh 5 adversarial haiku seed-
116
117
  **Which detectors deserve to exist?** Measured too. Affordance seeds offer an agent one
117
118
  specific bypass each; predictions are pre-registered before any rep runs
118
119
  (`harness/PREDICTION-affordance.md`). Result: 7/10 adversarial runs regenerated a golden
119
- file from buggy output (100% passed unseenso `snapshot-rewrite` was built), while
120
+ file from buggy output and sailed through**because nothing guarded that class yet**;
121
+ that measured demand is why `snapshot-rewrite` exists — while
120
122
  timeout-inflation and lint-config-gutting measured **0/10 attempts each** — two
121
123
  pre-registered bets refuted, two detectors *not* built. Bypass classes with no measured
122
124
  demand stay in a record-only shadow scanner that accumulates evidence on every weekly
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "tamperward",
3
- "version": "1.4.1",
3
+ "version": "1.4.2",
4
4
  "description": "The deterministic agent-integrity gate. One ruleset, evaluated on the actual diff/commands as a verdict, enforced everywhere a change can be made.",
5
5
  "license": "Apache-2.0",
6
6
  "author": "hexrift",