mirror of
https://github.com/deepseek-ai/deepseek-harness
synced 2026-08-15 21:04:50 +00:00
a9a44fea96
# Conflicts: # .agents/notes/implemented/architecture/2026-07-19-cooperative-tool-cancellation.i18n.yaml # .agents/notes/implemented/architecture/2026-07-19-cooperative-tool-cancellation.md # .agents/notes/implemented/architecture/2026-07-19-cooperative-tool-cancellation.zh.md # .agents/notes/proposed/architecture/2026-07-19-required-cancellation-through-tool-capability-seams.i18n.yaml # .agents/notes/proposed/architecture/2026-07-19-required-cancellation-through-tool-capability-seams.md # .agents/notes/proposed/architecture/2026-07-19-required-cancellation-through-tool-capability-seams.zh.md # docs/cookbook/adding-a-tool.i18n.yaml # docs/rfc/INDEX.md # packages/AGENTS.md
guard/ — loop-hygiene guard family
Behavioral guard plugins that watch the agent loop for unproductive patterns and nudge the model back on course. A single product package — there is no interface/implementation seam here, because a guard is a self-contained consumer of existing core seams (tools/post-execute, agent/prompt-submit, agent/status), not a swappable capability.
| Package | Role | ctx key |
|---|---|---|
repeat-tool-guard/ |
Advisory reminders when an agent loops on identical tool calls | (listens on ctx.tools' waterfalls) |
Reminders travel as additionalContexts on the tools/post-execute decision; the agent loop appends them as logged context/message events after the step's tool results (see the tools package), so everything a guard says to the model is reconstructable from the session log.