Skip to content

Ondra add agent input checking#486

Merged
ondrej-lukas merged 5 commits intomainfrom
ondra-add-agent-input-checking
Mar 10, 2026
Merged

Ondra add agent input checking#486
ondrej-lukas merged 5 commits intomainfrom
ondra-add-agent-input-checking

Conversation

@ondrej-lukas
Copy link
Member

@ondrej-lukas ondrej-lukas commented Mar 10, 2026

Fixes stability in the processing of input from agents:

  • sanitization of agent name (used when storing the trajectories server-side)
  • better handling of malformed actions (processing, reporting)
  • tests for action processing in the agent server/coordinator

@ondrej-lukas ondrej-lukas added this to the netsecgame v 0.1.1 milestone Mar 10, 2026
@ondrej-lukas ondrej-lukas linked an issue Mar 10, 2026 that may be closed by this pull request
[Bug]: Socket attribute typo in BaseAgent.__init__ causes silent broken-socket state #467
Closed
@ondrej-lukas ondrej-lukas merged commit f50feb3 into main Mar 10, 2026
2 checks passed
@ondrej-lukas ondrej-lukas deleted the ondra-add-agent-input-checking branch March 10, 2026 09:01
This was linked to issues Mar 10, 2026
Critical DoS: malformed action crashes run_game (UnboundLocalError) #475
Closed
[Bug]: Vulnerability by gpt-5.3-codex xhigh (experiment) #472
Closed
Critical DoS: malformed input dereference in run_game action dispatch #477
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment