Directories
¶
| Path | Synopsis |
|---|---|
|
Example: build a headless attach-mode daemon from the library — the canonical post-#388 embedding shape.
|
Example: build a headless attach-mode daemon from the library — the canonical post-#388 embedding shape. |
|
Example: drive an agent through autonomous.Run end-to-end with no LLM credentials.
|
Example: drive an agent through autonomous.Run end-to-end with no LLM credentials. |
|
Example: drive an autonomous run via the AutonomousHandle API.
|
Example: drive an autonomous run via the AutonomousHandle API. |
|
Example: drive an agent through autonomous.Run, simulate a crash via a tight max-turns budget, then continue the run with autonomous.Resume against the same SQLite event log.
|
Example: drive an agent through autonomous.Run, simulate a crash via a tight max-turns budget, then continue the run with autonomous.Resume against the same SQLite event log. |
|
Example: wire a background.Manager (built with background.NewManager and attached to the parent via agent.WithBackgroundManager) and demonstrate the in-process spawn pathway end-to-end with no LLM credentials.
|
Example: wire a background.Manager (built with background.NewManager and attached to the parent via agent.WithBackgroundManager) and demonstrate the in-process spawn pathway end-to-end with no LLM credentials. |
|
Example: minimal multi-turn agent using the core-agent library directly (no CLI).
|
Example: minimal multi-turn agent using the core-agent library directly (no CLI). |
|
Example: assemble a multi-session attach daemon from pkg/compose — the #386 story.
|
Example: assemble a multi-session attach daemon from pkg/compose — the #386 story. |
|
Example: one assistant turn fans out three independent background subagents via the spawn_agent tool family, and their reports drain back into the parent's next turn — all with no LLM credentials.
|
Example: one assistant turn fans out three independent background subagents via the spawn_agent tool family, and their reports drain back into the parent's next turn — all with no LLM credentials. |
|
Example: drive the agent loop offline by replaying a recorded JSONL transcript through the mock "scripted" provider — no LLM credentials required.
|
Example: drive the agent loop offline by replaying a recorded JSONL transcript through the mock "scripted" provider — no LLM credentials required. |
|
Example: the supervision-tree topology from docs/scheduled-monitoring-design.md, end-to-end with no LLM credentials.
|
Example: the supervision-tree topology from docs/scheduled-monitoring-design.md, end-to-end with no LLM credentials. |
|
Example: drive the agent with the standard built-in tools (read_file, list_dir, bash, etc.) and render the response as an interactive chat session via runner.WriteEvents.
|
Example: drive the agent with the standard built-in tools (read_file, list_dir, bash, etc.) and render the response as an interactive chat session via runner.WriteEvents. |
|
Example: drive a parent agent with a subagent end-to-end with no LLM credentials.
|
Example: drive a parent agent with a subagent end-to-end with no LLM credentials. |
|
Example: an agent with one custom tool, MCP servers loaded from .agents/mcp.json (if present), and SKILL.md skills from .agents/skills/ (if any).
|
Example: an agent with one custom tool, MCP servers loaded from .agents/mcp.json (if present), and SKILL.md skills from .agents/skills/ (if any). |
Click to show internal directories.
Click to hide internal directories.