fs: add /help ctl command — returns command reference

README.md

@@ -53,11 +53,9 @@ Everything lives under `$XDG_CONFIG_HOME/ollie/` (default: `~/.config/ollie/`)
 
 | Capability | How |
 |---|---|
-| **CLI** | `o` — unified namespace CLI: `read`, `write`, `ls`, `ctl`, `prompt`, `tui`, `env`, `generate`, `complete`, `route` |
+| **CLI** | `o` — unified namespace CLI: `read`, `write`, `ls`, `ctl`, `prompt`, `tui`, `env`, `generate` |
 | **Terminal TUI** | `o tui` — tmux layout, no widgets |
 | **One-shot LLM** | `o generate "explain monads"` or `echo "prompt" \| o generate` |
-| **Code completion** | `o complete '{"file":"main.go","prefix":"func "}'` or `echo '{"file":"..."}' \| o complete` |
-| **Task routing** | `o route "implement login"` or `echo "task" \| o route` → `backend=X model=Y` |
 | **Run an agent** | Create a session + agent via 9P — then connect with any frontend ([ellie](doc/ellie.md), KDE (GUI/KRunner/Kate), `o tui`) |
 | **Remote execution** | Set `remote=user@host` in session config |
 | **Agents** | Write a JSON config in `data/agents/` with agent prompt in `data/prompts/` |
@@ -98,7 +96,7 @@ graph TB
         TUI["o tui<br>(tmux + o read/ctl)"]
         KDE[KDE GUI<br>ollie-9p native client]
         KATE[Kate plugin]
-        EMACS[Emacs<br>ellie.el]
+        EMACS[Emacs<br>ellie.el · ollie-9p]
         ACME[Plan 9 acme]
         SCRIPTS[Shell scripts<br>o read · o ctl · o write]
     end
@@ -109,9 +107,8 @@ graph TB
 
     subgraph "olliesrv"
         P9[9P Filesystem<br>session/ namespace]
-        SM[Session Manager]
+        SESS[session.Session<br>tool server conn · persistence]
         AG[agent.Agent<br>loop · compaction<br>prompt assembly]
-        ROUTE[Router<br>complete · generate · route]
         TOOLS[toolsrv.Server<br>dynamic dispatch · sandbox]
     end
 
@@ -124,22 +121,19 @@ graph TB
     end
 
     subgraph Execution
-        LOCAL[Tool Scripts<br>loaded via ctl<br>· shell, reasoning_think<br>· file_*, lsp_*, memory_*<br>· gui_*, subagent_*]
+        EXEC[Tool Scripts<br>loaded via ctl<br>· shell, reasoning_think<br>· file_*, lsp_*, memory_*<br>· gui_*, subagent_*]
-        REMOTE[ollie-remote<br>via SSH · no embedded tools]
     end
 
-    TUI & SCRIPTS & EMACS & ACME --> O
+    TUI & SCRIPTS & ACME --> O
     O --> P9
-    KDE & KATE --> P9
+    KDE & KATE & EMACS --> P9
-    P9 --> SM
+    P9 --> SESS
-    SM --> AG
+    SESS --> AG
-    AG --> ROUTE
+    SESS -->|shared RPC conn| TOOLS
-    AG --> TOOLS
 
     AG -->|streaming| OLL & OAI & ANT & COP & KIRO
 
-    TOOLS --> LOCAL
+    TOOLS --> EXEC
-    TOOLS -->|SSH| REMOTE
 ```
 
 `olliesrv` exposes the agent runtime over a single surface: a 9P2000 filesystem.

contrib/elisp/el/ellie.el

@@ -614,15 +614,15 @@ Attaches to the last session if it still exists; otherwise creates a new one."
 
 (defun ellie--complete-send (prefix suffix)
   "Send a completion request with PREFIX and SUFFIX text.
-Returns the raw response from the server."
+Returns the raw response from the server via /generate."
-  (let ((payload (format "%s\n__COMPLETE_SEPARATOR__\n%s" prefix suffix)))
+  (let ((payload (format "Complete the code between the markers. Return ONLY the completion, no explanation.\n\n```\n%s<CURSOR>%s\n```" prefix suffix)))
     (with-temp-buffer
       (insert payload)
       (let ((rc (call-process-region (point-min) (point-max)
-                                     "ollie-9p" t t nil "rdwr" "complete")))
+                                     "ollie-9p" t t nil "rdwr" "generate")))
         (if (zerop rc)
             (buffer-string)
-          (error "ollie-9p complete failed (exit %d)" rc))))))
+          (error "ollie-9p generate failed (exit %d)" rc))))))
 
 (defun ellie-complete-at-point (&optional prefix suffix)
   "Request a completion for the region around point.

data/scripts/o

@@ -27,7 +27,7 @@ set -euo pipefail
 # ── Path classification ─────────────────────────────────────────
 
 # Root-level entries (no context needed)
-ROOT_ENTRIES="session bypass agents backends models help ctl eventwait complete generate route aliases tools"
+ROOT_ENTRIES="session bypass agents backends models help ctl eventwait generate aliases tools"
 
 # Session-level files (exist directly under session/{name}/)
 SESSION_FILES="env paused connected ctl name id bypass"
@@ -235,40 +235,6 @@ cmd_generate() {
     printf '%s' "$prompt" | ollie-9p rdwr generate
 }
 
-cmd_complete() {
-    # Code completion via /complete (no session/agent needed)
-    # Usage: o complete '{"file":"...","prefix":"...","suffix":"..."}'
-    #        echo '{"file":"..."}' | o complete
-    local json=""
-
-    if [[ $# -gt 0 ]]; then
-        json="$*"
-    else
-        json=$(cat)
-    fi
-
-    [[ -z "$json" ]] && { echo "usage: o complete <json>" >&2; exit 1; }
-
-    printf '%s' "$json" | ollie-9p rdwr complete
-}
-
-cmd_route() {
-    # Task routing via /route (no session/agent needed)
-    # Usage: o route "task description"
-    #        echo "task" | o route
-    local task=""
-
-    if [[ $# -gt 0 ]]; then
-        task="$*"
-    else
-        task=$(cat)
-    fi
-
-    [[ -z "$task" ]] && { echo "usage: o route <task>" >&2; exit 1; }
-
-    printf '%s' "$task" | ollie-9p rdwr route
-}
-
 cmd_ctl() {
     if [[ $# -lt 1 ]]; then
         echo "usage: o ctl <command> [args...]" >&2
@@ -518,10 +484,8 @@ Commands:
   ls [path]            List namespace entries
 
   generate <prompt>    One-shot generation (no session needed)
-  complete <json>      Code completion (no session needed)
-  route <task>         Route task to best model (no session needed)
 
-  ctl <cmd> [s] [a]   Send control command (session or agent level)
+  ctl <cmd> [args]     Send control command (rdwr: gets response)
 
   prompt               Interactive multi-line prompt REPL
   log [-n N]           Show last N lines of agent log (default 1000)
@@ -562,8 +526,6 @@ case "$subcmd" in
     write)    cmd_write "$@" ;;
     ls)       cmd_ls "$@" ;;
     generate) cmd_generate "$@" ;;
-    complete) cmd_complete "$@" ;;
-    route)    cmd_route "$@" ;;
     ctl)      cmd_ctl "$@" ;;
     prompt)   cmd_prompt "$@" ;;
     log)      cmd_log "$@" ;;

doc/resources/9p.md

@@ -64,7 +64,7 @@ for s in session/*/agent/*/prompt; do echo "run tests" > "$s"; done
 for s in session/*/agent/*/statewait; do cat "$s" > /dev/null; done
 
 # Grep all agent plans
-grep -r "TODO" session/*/plan
+grep -r "TODO" session/*/agent/*/plan
 
 # Monitor costs
 cat session/*/agent/*/cost
@@ -78,7 +78,7 @@ with a 10-line shell script instead of a framework.
 
 ### Plan 9 patterns
 
-- **`/net/dns` pattern** — `/complete`, `/generate`, `/route` are stateless
+- **`/net/dns` pattern** — `/generate` and `ctl` are stateless
   per-fid: write a request, read back the result
 - **`ctl` files** — control operations (stop, kill, rename, compact) are writes
   to a control file, not method calls
@@ -95,21 +95,6 @@ cat main.go | ollie-9p rdwr generate
 echo '{"prompt":"explain recursion"}' | ollie-9p rdwr generate
 ```
 
-### `/complete` — code completion
-
-Fill-in-the-middle:
-
-```sh
-echo '{"file":"main.go","prefix":"func ","suffix":""}' | ollie-9p rdwr complete
-```
-
-### `/route` — task routing
-
-```sh
-echo 'implement login page' | ollie-9p rdwr route
-# backend=anthropic model=claude-sonnet-4-20250514
-```
-
 ### Key files per agent
 
 | File | Mode | Description |

doc/resources/architecture.md

@@ -205,7 +205,6 @@ Agent configs declare prompts as a JSON array of shell commands. Each command is
     ├── {name}/
     │   ├── env            session environment variables
     │   ├── ctl            session control (rdwr: kill, save, invalidate, pause, resume)
-    │   ├── plan           session-scoped markdown checklist
     │   ├── id             immutable session UUID
     │   ├── name           mutable session name (write to rename)
     │   ├── bypass        per-session bypass policy

doc/resources/multi-agent.md

@@ -15,12 +15,12 @@ to its `name` file moves the entire directory subtree.
 |------|---------|
 | `/session/new` | Create an empty session (write `name=...`, read back name) |
 | `/session/{name}/agent/new` | Create an agent within a session (write `cwd=... backend=...`) |
-| `/session/{name}/plan` | Session-scoped markdown checklist |
 | `/session/{name}/env` | Session environment variables |
 | `/session/{name}/id` | Immutable session UUID |
 | `/session/{name}/name` | Mutable session name (write to rename) |
 | `/session/{name}/agent/{aname}/prompt` | Submit a prompt |
-| `/session/{name}/agent/{aname}/chat` | Streaming chat log (blocking read) |
+| `/session/{name}/agent/{aname}/chat` | Streaming chat (filtered text) |
+| `/session/{name}/agent/{aname}/plan` | Agent-scoped markdown checklist |
 | `/session/{name}/agent/{aname}/cfg` | Agent config KV: backend, model, agent, cwd, params |
 | `/session/{name}/agent/{aname}/statewait` | **Blocking read** — returns when state changes |
 | `/session/{name}/agent/{aname}/ctl` | Control commands: `stop`, `kill`, `compact`, `rn <name>` |
@@ -32,16 +32,14 @@ Global request-response files provide stateless operations without sessions:
 | File | Purpose |
 |------|---------|
 | `/generate` | Write prompt, read response (one-shot generation) |
-| `/complete` | Write JSON `{file, prefix, suffix}`, read code completion |
-| `/route` | Write task description, read `backend=X model=Y` recommendation |
 
 Global namespaces are shared across all sessions:
 
 | Path | Purpose |
 |------|---------|
 | `/session/idx` | Session index (id, state, cwd, backend, model) |
-| `/session/{name}/plan` | Shared session-scoped checklist |
+| `/session/{name}/agent/{aname}/plan` | Agent checklist |
-| `/session/{name}/agent/{aname}/chat` | Full chat history (searchable) |
+| `/session/{name}/agent/{aname}/chat` | Streaming filtered text |
 
 ## Spawning Agents
 
@@ -84,7 +82,7 @@ flowchart TB
 ```
 
 A root agent reasons about a task and delegates subtasks to workers. Workers
-report back via their chat history; the conductor reads `/session/{name}/agent/{aname}/chat` after
+report back via their log; the conductor reads `/session/{name}/agent/{aname}/log` after
 `statewait` unblocks.
 
 ```sh

doc/usage.md

@@ -71,8 +71,6 @@ o read chat    # stream chat output
 | `o ls [path]` | List namespace entries |
 | `o ctl <cmd> [args]` | Send control command (rdwr: gets response) |
 | `o generate <prompt>` | One-shot generation (no session needed) |
-| `o complete <json>` | Code completion |
-| `o route <task>` | Route task to best model |
 | `o tui` | Launch tmux TUI |
 | `o env [session] [agent]` | Print export statements for context |
 

fs/handlers.go

@@ -774,6 +774,23 @@ var agentCtlHandlers = map[string]rdwrHandler{
 	"systemprompt": func(ctx HandlerCtx, _ []string) ([]byte, error) {
 		return []byte(ctx.Agent.SystemPrompt() + "\n"), nil
 	},
+	"help": func(_ HandlerCtx, _ []string) ([]byte, error) {
+		return []byte(`stop           — interrupt running agent
+compact        — compact context
+clear          — clear history
+inject <text>  — inject prompt mid-turn (overwrites pending)
+agent [name]   — show/switch agent profile
+model [name]   — show/switch model
+models         — list available models
+tools          — list loaded tools
+tool_load <n>  — load a tool
+cwd [path]     — show/change working directory
+name [name]    — show/rename agent
+backend        — show backend
+systemprompt   — show rendered system prompt
+help           — this message
+`), nil
+	},
 }
 
 func readAgentStats(ctx HandlerCtx) ([]byte, error) {

kde

@@ -1 +1 @@
-Subproject commit ec3419aa4e710eb63d449be612995a4e4b6f46ca
+Subproject commit c6da56310992d19049a60910955ab2b915a9e5aa

prompts/system_prompt.md

@@ -165,7 +165,7 @@ echo '{"file":"main.go","prefix":"func "}' | ollie-9p rdwr complete
 
 | File | Mode | Purpose |
 |---|---|---|
-| `plan`        | r/w   | Session-scoped markdown checklist; survives compaction |
+| `plan`        | r/w   | Agent-scoped markdown checklist; survives compaction |
 | `env`         | read  | Session environment variables |
 | `agent/`      | dir   | Agents within this session |