CrazyBoyM bb232a5316 fix(v2): inject NAG_REMINDER inside agent_loop, not between tasks ...

Previously NAG_REMINDER was only checked in main() after a task completed,
meaning models never saw the reminder during long-running tasks.

Now NAG_REMINDER is injected inside agent_loop when rounds_without_todo > 10,
so models see it during task execution (matching Kode's behavior).

Fixes #14

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

2026-01-25 00:15:12 +08:00

..

v0-bash-is-all-you-need.md

Add navigation link to v1 model documentation

2026-01-07 11:07:28 +05:30

v0-Bash就是一切.md

Complete rewrite: original educational content only

2025-12-31 07:01:42 +08:00

v1-model-as-agent.md

Update navigation links in v1-model-as-agent.md

2026-01-07 11:08:15 +05:30

v1-模型即代理.md

Complete rewrite: original educational content only

2025-12-31 07:01:42 +08:00

v2-structured-planning.md

fix(v2): inject NAG_REMINDER inside agent_loop, not between tasks

2026-01-25 00:15:12 +08:00

v2-结构化规划.md

Complete rewrite: original educational content only

2025-12-31 07:01:42 +08:00

v3-subagent-mechanism.md

Update navigation links in v3-subagent-mechanism.md

2026-01-02 15:05:45 +05:30

v3-子代理机制.md

Complete rewrite: original educational content only

2025-12-31 07:01:42 +08:00

v4-skills-mechanism.md

Add navigation links to skills mechanism documentation

2026-01-02 15:08:38 +05:30

v4-Skills机制.md

Complete rewrite: original educational content only

2025-12-31 07:01:42 +08:00