qzl/social-app
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: 实现 AgentScope tool call context，支持 runtime 上下文续接

Browse Source
This commit is contained in:
qzl
2026-03-17 14:12:44 +08:00
parent d6cc5a0dae
commit 3bf7640000
15 changed files with 547 additions and 243 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/AGENTS.md
+5 -3
View File
@@ -251,10 +251,12 @@ Agent loop functionality MUST follow the AG-UI protocol. **Use the `ag-ui` skill
Custom tool `ToolAgentOutput` MUST follow these rules:
- Use field name `result` only. Do not introduce or keep `result_summary` compatibility aliases.
- `metadata.tool_agent_output` is the canonical source for runtime observation and history replay.
- `tool_call_args` stores input snapshot only; avoid mixing execution output into `tool_call_args`.
- `result` stores output facts only; do not repeat input parameters already present in `tool_call_args`.
- `result` is for downstream agent reasoning and tool chaining, not for end-user presentation.
- Prefer compact structural facts over prose: include identifiers and execution-critical facts (`id`, `status`, `count`, `page`, operation outcome, missing required args).
- For list/read tools, include multiple candidate records when needed (at least top matches) with stable identifiers.
- For write tools, always include affected resource identifiers in `result`.
- For list/read tools, include multiple candidate records when needed (at least top matches) with stable identifiers and scheduling-critical fields.
- For write tools, include per-item operation outcomes and affected resource identifiers in `result`.
- Keep `result` concise, deterministic, and machine-oriented; avoid decorative wording and UI-style formatting.
## Multi-Agent Orchestration (AgentScope Framework)