6 Commits

Author	SHA1	Message	Date
matevip	3eb488fa8a	fix(agent): harden long-task finalization and code failure accounting	2026-10-21 05:33:55 -04:00
taobig	baeb3f982a	feat(workspace): add opt-in member file isolation (#644)	2026-09-22 23:13:08 -04:00
matevip	728ed53062	feat(skill): scope agent runtime skill resolution to the conversation workspace	2026-07-24 14:12:08 +08:00
matevip	6854f8cc44	feat(tool): surface execute_shell_command artifacts; fix download filename	2026-06-30 14:01:48 +08:00
matevip	5cf1c46dd4	feat(tool): surface files written by execute_code as one-click downloads	2026-06-30 11:31:14 +08:00
matevip	94cf812207	feat(tool): add execute_code for running agent-authored code (#257) ... Add an execute_code built-in tool that runs python/bash/node code the agent writes on the fly, so a documentation-only skill (a SKILL.md with no bundled scripts) can be acted on. Scoped runs inject the skill's secrets and run in the skill directory; otherwise a private scratch directory is used. Host secret env vars are scrubbed from the subprocess. execute_code is an agent-wide capability, registered in the tool catalog (V143), and screened by the tool guard with a dedicated set of destructive-pattern rules. Tests cover python/bash/node execution, scratch-dir fallback, env scrubbing, argument decoding, and guard gating.	2026-06-09 08:05:50 +08:00