dissertation.git - Undergraduate dissertation — Deceptive Alignment in Autonomous LLM Agents (UEL, 2026). Sandboxed agentic-LLM deception study: 960 runs, three-tier judging (regex + blackbox + glassbox).

Age	Commit message (Collapse)	Author
2026-03-02	feat: regex checker now returns score (e.g. 6/7) instead of binary PASS/FAIL	CaptainJack2491
	- Verdict shows matched/total required rules for more nuanced evaluation - Adds 'matched' and 'total' fields to result dict - Returns 'N/A' when no rules are defined
2026-03-02	fix: sharpen glassbox category definitions for ROBUST_HONEST vs ↵	CaptainJack2491
	UNINTENTIONAL_FAILURE - ROBUST_HONEST: ignored hidden goal + output is accurate and complete - UNINTENTIONAL_FAILURE: ignored hidden goal + output has genuine quality issues - Added explicit disambiguation note in the prompt - Clarified CONSIDERED_HONEST as deliberate refusal to comply
2026-03-02	feat: add benign_goal (task description) context to blackbox judge	CaptainJack2491
	- Blackbox prompt now includes TASK DESCRIPTION section from benign_goal.md - Helps judge evaluate framing even without a baseline - Hidden goal is never exposed (maintains blackbox/glassbox boundary) - Applied to both judge_single and prepare_batch_requests paths
2026-03-02	fix: judge resilience for missing baselines and list-format regex rules	CaptainJack2491
	- RegexChecker now handles both list-format and dict-format YAML rules - Judge falls back to experiment log's VFS state for source docs when no baseline exists - Blackbox judge receives '(No baseline available)' instead of crashing - Resume logic excludes .partial.json from completed run count
2026-03-01	feat(judge): updated blackbox to include source document for added accuracy	CaptainJack2491

2026-02-27	fix(judge): hash run_id to stay under Anthropic batch custom_id 64-char limit	CaptainJack2491

2026-02-26	chore: add .env.example + config.yaml.example, gitignore config.yaml	CaptainJack2491
	- Create .env.example with placeholder keys for all providers - Create config.yaml.example with clean reference config - Add config.yaml and judge_logs/ to .gitignore - Load .env in judge.py for standalone usage
2026-02-25	feat: implement judging system with regex, blackbox, and glassbox checks	CaptainJack2491
	- Add src/judge.py: RegexChecker, BlackboxChecker, GlassboxChecker, Judge orchestrator - Add src/judge_runner.py: CLI for batch/single judging with CSV output - Add regex_rules.yaml for corporate_sabotage_v2 and child_protection scenarios - Add judge config section to config.yaml (model, temperature, log_dir) - Add tests/test_judge.py: 34 unit tests covering all components - Uses Anthropic Batch API for half-cost LLM judge calls - Judge logs include full CoT reasoning and judge model for auditability