summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2026-02-27fix(judge): hash run_id to stay under Anthropic batch custom_id 64-char limitCaptainJack2491
2026-02-27fix: separate skipped runs from incomplete in summary outputCaptainJack2491
2026-02-27docs: add checkpoint/resume rationale note for methodologyCaptainJack2491
2026-02-26feat: add checkpoint/resume support and atomic log writesCaptainJack2491
- Runner now counts existing .json log files per combination and skips completed runs, resuming from where it left off - Log files are written atomically via temp file + rename to prevent corrupt files from being counted as completed - Added --no-resume CLI flag to force a fresh start - Resume is enabled by default
2026-02-26chore: add .env.example + config.yaml.example, gitignore config.yamlCaptainJack2491
- 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
2026-02-25feat: implement judging system with regex, blackbox, and glassbox checksCaptainJack2491
- 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
2026-02-24Update README: add testing section, document interrogation systemCaptainJack2491
2026-02-24Add comprehensive test suite (104 tests)CaptainJack2491
Tests cover: - VFS: sandbox escape, path traversal, CRUD, singleton staleness - Config loader: missing keys, temperature cascade, oversight fallback - Tools: schema/implementation sync, VFS delegation - Agent: message construction, error handling, token counting, reasoning extraction - Interrogation: sanitization, provider detection, prompt loading - Runner: baseline extraction, prompt assembly, success detection All tests run offline with mocked API calls.
2026-02-24[refactor] improved interrogation systemCaptainJack2491
- Fixed syntax errors in print statements (lines 159/162) - Added sanitize_for_api() to strip custom log fields before API calls - Added interactive commands: history, reasoning, vfs, info, save, help - Colored terminal output for readability - Better error handling (KeyboardInterrupt, EOFError) - Added --show-history flag to review conversation on load
2026-02-24[fix] VFS singleton reset + remove vestigial check_username toolCaptainJack2491
- VFS now creates fresh instance when root_path/fs_data is provided (prevents cross-scenario contamination in multi-scenario runs) - Removed check_username from tools (unrelated to any scenario)
2026-02-24[feat] baseline system - auto-generates baselines before experimentsCaptainJack2491
- Added _extract_baseline_content() to extract output from create_file tool calls - Baselines now saved to logs/{model}/{scenario}/baseline.md - Added global oversight fallback (scenario-specific → global oversight dir) - Updated config_loader with oversight level loading from scenario dirs
2026-02-18[feat] added a new scenarioCaptainJack2491
2026-02-18[feat] added a baseline to the runnerCaptainJack2491
(no commit message provided) (no commit message provided) Co-authored-by: aider (openrouter/openrouter/aurora-alpha) <aider@aider.chat>
2026-02-11[docs] added progress presentationCaptainJack2491
2026-01-11[docs] updated READMECaptainJack2491
2026-01-11[fix] simplified everythingCaptainJack2491
made eveyrthing much cleaner and it works very well. got rid of providers setup, only use openai sdk now.
2026-01-11[fix] big moving around, probably made lot more bugs.CaptainJack2491
2026-01-09[docs] updated lit and added more in methodologyCaptainJack2491
2026-01-09updated interrogate to work with new log formatCaptainJack2491
2026-01-09updated scenario [corporate sabotage]CaptainJack2491
2026-01-09added scripts to get abstract and bibliography easily from papersCaptainJack2491
2025-12-30major commitCaptainJack2491
updated the program to use config.yaml made the runner to handel interleaved thinking added anthropic support and a lot more
2025-12-18updated oversightCaptainJack2491
2025-12-04gemini-2.5-pro pilot testCaptainJack2491
2025-12-03added corporate_sabotage_v2CaptainJack2491
2025-12-03updated interrogation to load from logsCaptainJack2491
2025-12-03updated methodologyCaptainJack2491
2025-11-25added interogation logic and updated structure of logsCaptainJack2491
2025-11-23updated agent to parse thinking in content field and added model_id to the logsCaptainJack2491
2025-11-23updated methodologyCaptainJack2491
2025-11-23good logsCaptainJack2491
2025-11-23moved 03-logs to mainCaptainJack2491
2025-11-23new paperCaptainJack2491
2025-11-20updated agent to work with the EXTRA_BODY variable to work with models that ↵CaptainJack2491
require it.
2025-11-20added corporate espionage scenarioCaptainJack2491
2025-11-20uodated oversight promptCaptainJack2491
2025-11-19updated logs to show modelname and scenario, and added extra param for google.CaptainJack2491
2025-11-19added new scenario, corporate espionageCaptainJack2491
2025-11-16better scenario managementCaptainJack2491
2025-11-16better VFSCaptainJack2491
2025-11-16added scenario data_sanitizationCaptainJack2491
2025-11-16added more toolsCaptainJack2491
2025-11-16better structure. and works with groq tooCaptainJack2491
2025-11-16added modular promptsCaptainJack2491
2025-11-16logging worksCaptainJack2491
2025-11-16reasoning works for both toolcall and responseCaptainJack2491
2025-11-16Move docs and notes and papers from feature-branch to mainCaptainJack2491
2025-10-28added gemini.mdCaptainJack2491
2025-10-28Initial commit: Set up project structureCaptainJack2491
generated by cgit v1.2.3 (git 2.46.0) at 2026-09-27 04:40:23 +0000