Issues

• Docs: remaining work after #670

  documentation
  AISecurityLab/hackagent#676 · Filter by author franconicola opened on Oct 9, 2026
  

• Define one precedence for API keys

  enhancement
  AISecurityLab/hackagent#673 · Filter by author franconicola opened on Oct 9, 2026

• Store the judge panel audit with the run

  enhancement
  AISecurityLab/hackagent#672 · Filter by author franconicola opened on Oct 9, 2026
  1

  

• Campaign runs: align what local and remote storage record

  bug
  AISecurityLab/hackagent#671 · Filter by author franconicola opened on Oct 9, 2026
  1

• Support self-hosted API endpoints: base URL with path prefix and custom CA bundle

  enhancement
  AISecurityLab/hackagent#669 · Filter by author franconicola opened on Oct 5, 2026

• Codex Trusted Directory Failure

  AISecurityLab/hackagent#657 · Filter by author hassen8 opened on Sep 25, 2026
  1

• Hermes provider fails: stdin vs -z PROMPT + --source argument doesn't exist in top level + missing binary="ollama" method of running

  AISecurityLab/hackagent#656 · Filter by author hassen8 opened on Sep 25, 2026
  1

• Token Counts Redacted by Tracking Sanitizer

  bug
  Bug

  AISecurityLab/hackagent#642 · Filter by author hassen8 opened on Sep 23, 2026
  2

  

• feat(attacks): Plan-of-Thought (PoT) backdoor (ASB)

  AISecurityLab/hackagent#602 · Filter by author franconicola opened on Sep 20, 2026

• feat(attacks): AgentPoison-style memory/RAG backdoor triggers

  AISecurityLab/hackagent#601 · Filter by author franconicola opened on Sep 20, 2026

• feat(attacks): STAC — sequential tool attack chaining (multi-turn)

  AISecurityLab/hackagent#600 · Filter by author franconicola opened on Sep 20, 2026

• feat(attacks): ToolHijacker — poison tool docs to hijack tool selection

  AISecurityLab/hackagent#599 · Filter by author franconicola opened on Sep 20, 2026

• feat(evaluation): add human-annotated benchmark for judge reliability

  enhancement
  Feature

  AISecurityLab/hackagent#586 · Filter by author franconicola opened on Sep 16, 2026 · v1
  

• Reporting of attack performance for GPAI models

  AISecurityLab/hackagent#433 · Filter by author franconicola opened on Jun 12, 2026

• Autoscan with LLM

  AISecurityLab/hackagent#355 · Filter by author marcorusso97 opened on May 14, 2026