Files
timmy-config/training/data
Alexander Payne 837e42d4ea
Some checks failed
Smoke Test / smoke (pull_request) Failing after 22s
Architecture Lint / Linter Tests (pull_request) Successful in 26s
Validate Config / YAML Lint (pull_request) Failing after 15s
Validate Config / JSON Validate (pull_request) Successful in 18s
Validate Config / Python Syntax & Import Check (pull_request) Failing after 56s
Validate Config / Python Test Suite (pull_request) Has been skipped
Validate Config / Cron Syntax Check (pull_request) Successful in 10s
Validate Config / Shell Script Lint (pull_request) Failing after 58s
Validate Config / Deploy Script Dry Run (pull_request) Successful in 10s
Validate Config / Playbook Schema Validation (pull_request) Successful in 24s
Validate Training Data / validate (pull_request) Successful in 22s
Architecture Lint / Lint Repository (pull_request) Failing after 15s
PR Checklist / pr-checklist (pull_request) Successful in 2m59s
feat(adversary): add jailbreak 1K prompt generator targeting SOUL.md values
- training/build_jailbreak_1000.py: generator producing 1000 prompts
- training/data/adversary/jailbreak_1000.jsonl: 1000 generated prompts
- tests/test_jailbreak_generator.py: 10 tests validating generator

Targets 6 SOUL.md value categories:
  deception (167), harm (167), identity (167), certainty (167),
  shutdown_resistance (166), crisis (166)

Crisis prompts include SOUL-compliant expected_behavior/failure_criteria.

Closes #578
2026-04-26 17:35:33 -04:00
..