pi-open-agents 0.1.17 → 0.1.19

package/README.md

@@ -13,6 +13,8 @@ Unified agent and subagent management for [pi](https://pi.dev), with [OpenCode](
 Replaces `pi-agent-mode` + `@johnnywu/pi-subagents` with one coherent plugin.
 
 > **Why pi?** Pi's minimalist core keeps system prompts under 1,000 tokens — making it exceptionally fast on local models and cheap on cloud APIs. pi-open-agents is built to leverage that minimalism. See [Pi vs OpenCode: Performance & Architecture](./PI-VS-OPencode.md) for a detailed comparison.
+>
+> 🐑 **Herd your agents.** pi-open-agents + [Herdr](https://herdr.dev) runs a full team of coding agents — persistent workspaces, live agent awareness, control from one screen. **[Set up the stack in 5 commands →](./PI-WITH-HERDR.md)**
 
 ---
 
@@ -82,6 +84,23 @@ have no permission system. `pi-open-agents` replaces both:
 
 ---
 
+## Popular requests, already covered
+
+Some of the most upvoted agent/subagent asks across the ecosystem are already
+here — or native in pi itself:
+
+| Asked elsewhere (👍) | Status in pi + pi-open-agents |
+|---|---|
+| `AGENTS.md` adoption — anthropics/claude-code#31005 👍336, openai/codex#12115 👍109 | ✅ Native in pi: `AGENTS.md`/`CLAUDE.md` discovery with `AGENTS.override.md` layering |
+| Agent Skills standard (`.agents/skills/`) — agentskills/agentskills#15 👍183 | ✅ Native in pi; shared agents from `.agents/` (#7 aligns to the dotagents layout) |
+| Tool restrictions that actually reach subagents — anthropics/claude-code#18950 👍70 | ✅ Since v0.1.14, `tools:`/`permission:` restrict the child toolset via `--tools` (full parent inheritance → #8) |
+| Model selection per subagent / per call — sst/opencode#6651 👍71 | ✅ Per-agent `model:` + per-call `model:` override in `subagent()` (since v0.1.19) |
+| Reasoning effort per agent — Kilo-Org/kilocode#13105 👍8 | ✅ `thinking: off…xhigh` per agent, forwarded to the child process |
+| Delegation guards — Kilo-Org/kilocode#9985 | ✅ `allowedAgents` allowlist + `maxDepth` recursion limit |
+| Agent management UI — openai/codex#22321 👍41 | ✅ `/agent`, `/agents`, `/agent-search`, `Ctrl+Shift+M`, programmatic `set_agent` |
+
+---
+
 ## Features
 
 ### Per-agent model, thinking, and permissions
@@ -135,7 +154,7 @@ If an agent has the `subagent` tool, the plugin automatically appends a
 subagents and the correct call syntax:
 
 ```
-subagent({ agent: "<name>", task: "<task>" })
+subagent({ agent: "<name>", task: "<task>", cwd?, session?, model? })
 ```
 
 You never need to explain delegation mechanics in agent prompts. The plugin
@@ -152,6 +171,11 @@ This means:
 - Each subagent runs with exactly the model and tools it declares
 - Skills load per-agent, with wildcard support (`security-*`, `git-*`)
 
+**Per-call model override:** the caller can override the agent's default for a
+single call — `subagent({ agent: "deep-research", task: "...", model: "zai/glm-5.3-flash" })`.
+Handy for "run this one task on the cheap model" without forking an agent
+definition. Malformed values (`no-slash`) are rejected before any spawn.
+
 The `--tools` whitelist is derived from both explicit `tools:` arrays and
 permission allow-lists. An agent with `permission: { read: allow, edit: allow }`
 will only get `read` and `edit` in the child process. Wildcard permissions
@@ -290,6 +314,24 @@ Optional cleanup:
 
 ---
 
+## Run a fleet 🐑
+
+Pi stays minimal on purpose — one agent, done well. When you need **many**,
+add [Herdr](https://herdr.dev): persistent workspaces, mouse-first UX, live
+agent awareness (`idle` / `working` / `blocked`), and cross-agent control
+from one screen — pi, or any mix of the 25 supported agent kinds. Six agents,
+six sessions, zero chaos: that's the exact workstation setup documented in
+**[pi × Herdr × pi-open-agents](./PI-WITH-HERDR.md)**.
+
+## Roadmap
+
+- **Parent permission inheritance** for subagents — #8
+- **Agent hot-reload** (or `/reload-agents`) — #9
+- ~~**Per-call model override** in `subagent()`~~ — ✅ shipped in [0.1.19](https://github.com/andrea-tomassi/pi-open-agents/releases/tag/v0.1.19) (#10)
+- **Parallel subagent limit** — #11
+
+---
+
 ## Development
 
 **Contributing?** Please read the [Contributing Guide](./CONTRIBUTING.md) before

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-open-agents",
-  "version": "0.1.17",
+  "version": "0.1.19",
   "description": "Unified agent and subagent management for pi coding agent, with OpenCode-compatible agent definitions.",
   "keywords": [
     "pi-package",

package/src/subagent/executor.ts

@@ -113,6 +113,8 @@ export interface RunSubagentOptions {
   outputArchiveDir?: string;
   agentDir?: string;
   session?: SubagentSessionInfo;
+  /** Per-call model override ("provider/model-id"). Takes precedence over agent.model. */
+  modelOverride?: string;
   resolvePi?: () => Promise<PiResolution> | PiResolution;
   runner?: ProcessRunner;
   fs?: ExecutorFs;
@@ -504,9 +506,10 @@ export async function runSubagent(options: RunSubagentOptions): Promise<AgentRes
       args.push("--no-context-files");
     }
 
-    // Model override
-    if (options.agent.model) {
-      args.push("--model", options.agent.model);
+    // Model: per-call override wins over the agent definition default
+    const modelId = options.modelOverride ?? options.agent.model;
+    if (modelId) {
+      args.push("--model", modelId);
     }
 
     // Thinking level

package/src/subagent/tool.ts

@@ -47,6 +47,7 @@ interface SubagentParams {
   task: string;
   cwd?: string;
   session?: SubagentSessionMode;
+  model?: string;
 }
 
 // ─── Helpers ─────────────────────────────────────────────────────────────────
@@ -199,7 +200,7 @@ export function registerSubagentTool(
     name: "subagent",
     label: "Subagent",
     description: "Delegate a task to a named sub-agent running in an isolated pi process.",
-    promptSnippet: "Delegate isolated tasks with subagent({ agent, task, cwd?, session? }).",
+    promptSnippet: "Delegate isolated tasks with subagent({ agent, task, cwd?, session?, model? }).",
     promptGuidelines,
     parameters: Type.Object({
       agent: Type.String({ description: "Name of the agent to invoke" }),
@@ -210,6 +211,11 @@ export function registerSubagentTool(
           description: 'Parent session handling: none starts a fresh subagent session, fork branches from the current session',
         }),
       ),
+      model: Type.Optional(
+        Type.String({
+          description: 'Per-call model override in "provider/model-id" format (e.g. "anthropic/claude-sonnet-4-5"). Takes precedence over the agent definition default.',
+        }),
+      ),
     }),
 
     async execute(
@@ -237,6 +243,13 @@ export function registerSubagentTool(
       }
 
       const childCwd = params.cwd ?? ctx.cwd;
+
+      if (params.model !== undefined && !/^[^/\s]+\/[^/\s]+$/.test(params.model)) {
+        throw new Error(
+          `Invalid model "${params.model}": expected "provider/model-id" format (e.g. "anthropic/claude-sonnet-4-5").`,
+        );
+      }
+
       const session = resolveSubagentSession(
         params.session,
         childCwd,
@@ -254,6 +267,7 @@ export function registerSubagentTool(
         availableAgents: availableSubagentsForAgent(agent, availableSubagents),
         agentDir: options.agentDir,
         session,
+        modelOverride: params.model,
         onProgress: (progress) => onUpdate?.(toProgressResult(progress)),
       });