Files
stack/packages/control-board/src/scan.mjs
T
jason.woltjeandClaude Fable 5.1 6ec253de20 Pin the mid-tool-call rule in the control board scanner (#1503)
Acceptance rule (plan page, bf641e22): a seat mid-tool-call is working,
never waiting. The scanner already met it through pi's stopReason values;
deriveState now also checks the content for a toolCall block (working),
after the error stop reasons and before "stop" (waiting). Thinking blocks
do not keep a text turn from being waiting. Three JSONL fixture tests and
three state-table cases pin the rule. Live check on the real board:
orch-01 and rev-code-01 mid-tool-call are working, velma's finished
text-only turn is waiting. Sonnet review: APPROVED.

Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-12 09:29:19 -05:00

276 lines
12 KiB
JavaScript

// Control board step 1: read each agent's newest pi session log and tmux
// liveness, and write one small JSON status file per agent.
//
// States (plain words):
// working - the agent is in the middle of a turn (thinking or running tools)
// waiting - the agent finished its turn; it is your move
// error - the agent's last turn ended in an error, was aborted, or was cut off; look at it
// offline - no tmux session for this agent, or its session no longer runs pi
// idle - the agent is live but has no conversation yet
// unknown - liveness could not be checked (tmux missing or unresponsive); not a guess
//
// Board files are derived and rewritable. They are not run records. The one
// exception is <boardDir>/seen.json, which holds Jason's "seen" marks and is
// only changed when he clicks; a scan reads it and never rewrites it.
import { existsSync, readFileSync, readdirSync, statSync, mkdirSync, writeFileSync, renameSync } from "node:fs";
import { join, basename, isAbsolute, resolve } from "node:path";
import { homedir } from "node:os";
import { spawnSync } from "node:child_process";
export const STATES = Object.freeze(["working", "waiting", "error", "offline", "idle", "unknown"]);
const TEXT_LIMIT = 240;
export class ConfigError extends Error {}
export function defaultConfigPath() {
return join(homedir(), ".config", "mosaic-dev", "config.json");
}
// Fail closed: the config must exist, parse, and name an absolute dataRoot.
export function loadConfig(path = defaultConfigPath()) {
if (!existsSync(path)) throw new ConfigError(`config not found: ${path}`);
let raw;
try {
raw = JSON.parse(readFileSync(path, "utf8"));
} catch (err) {
throw new ConfigError(`config is not valid JSON: ${path} (${err.message})`);
}
if (!raw || typeof raw !== "object" || Array.isArray(raw)) throw new ConfigError(`config is not an object: ${path}`);
if (typeof raw.dataRoot !== "string" || !isAbsolute(raw.dataRoot)) {
throw new ConfigError(`config.dataRoot must be an absolute path: ${path}`);
}
return { dataRoot: raw.dataRoot };
}
// Newest *.jsonl under a directory tree, by mtime. Returns null when none.
export function findNewestSession(dir) {
if (!existsSync(dir)) return null;
let best = null;
const walk = (d) => {
for (const name of readdirSync(d)) {
const p = join(d, name);
const st = statSync(p);
if (st.isDirectory()) walk(p);
else if (name.endsWith(".jsonl") && (!best || st.mtimeMs > best.mtimeMs)) best = { path: p, mtimeMs: st.mtimeMs };
}
};
walk(dir);
return best ? best.path : null;
}
function collapse(text) {
const one = text.replace(/\s+/g, " ").trim();
return one.length > TEXT_LIMIT ? one.slice(0, TEXT_LIMIT - 1) + "…" : one;
}
// Read a pi session log. A partially written final line is skipped and counted,
// not treated as fatal, because pi may be appending while we read.
export function readSession(file) {
const lines = readFileSync(file, "utf8").split("\n");
let sessionId = null, cwd = null, lastTimestamp = null, lastMessage = null, lastAssistantText = null, lastError = null;
let skippedLines = 0;
for (const line of lines) {
if (!line.trim()) continue;
let entry;
try {
entry = JSON.parse(line);
} catch {
skippedLines += 1;
continue;
}
if (entry.timestamp) lastTimestamp = entry.timestamp;
if (entry.type === "session") {
sessionId = entry.id ?? sessionId;
cwd = entry.cwd ?? cwd;
} else if (entry.type === "message" && entry.message) {
lastMessage = entry.message;
if (entry.message.role === "assistant" && Array.isArray(entry.message.content)) {
const text = entry.message.content.filter((c) => c && c.type === "text" && typeof c.text === "string").map((c) => c.text).join("\n");
if (text.trim()) lastAssistantText = collapse(text);
lastError = entry.message.stopReason === "error" && typeof entry.message.errorMessage === "string" ? collapse(entry.message.errorMessage) : null;
}
}
}
return { file, sessionId, cwd, lastTimestamp, lastMessage, lastAssistantText, lastError, skippedLines };
}
// True when an assistant message carries a tool call in its content.
export function hasToolCall(message) {
return Array.isArray(message?.content) && message.content.some((c) => c && c.type === "toolCall");
}
// Pure state rule. alive: true/false, or null when liveness could not be checked.
// A null check is reported as "unknown" rather than assumed alive (fail closed).
// Acceptance rule (plan page, 2026-09-12): a seat mid-tool-call is working,
// never waiting. The newest entry being an assistant message with a tool call,
// or a tool result, means working even if the last text looked like a question.
// waiting needs a text-only assistant message whose turn ended (stopReason stop).
export function deriveState({ alive, session }) {
if (alive === false) return "offline";
if (alive !== true) return "unknown";
if (!session || !session.lastMessage) return "idle";
const m = session.lastMessage;
if (m.role === "assistant") {
if (m.stopReason === "error" || m.stopReason === "aborted" || m.stopReason === "length") return "error";
if (hasToolCall(m)) return "working";
if (m.stopReason === "stop") return "waiting";
return "working";
}
return "working";
}
// Programs that count as a live pi agent in a tmux pane. A tmux session that
// still exists but only runs a shell (or another harness) is not alive: its
// pi session log is history, not status.
export const PI_COMMANDS = Object.freeze(["pi"]);
export function panesRunPi(listPanesOutput) {
return String(listPanesOutput)
.split("\n")
.map((l) => l.trim())
.some((cmd) => PI_COMMANDS.includes(cmd));
}
// true: a pane in the tmux session runs pi. false: no such session, or no pane
// runs pi. null: tmux could not be run at all (reported as "unknown", never
// assumed alive). `exec` is injectable for tests.
export function tmuxIsAlive({ socket, session }, { exec = spawnSync } = {}) {
const args = [];
if (socket) args.push("-L", socket);
args.push("list-panes", "-s", "-t", `=${session}`, "-F", "#{pane_current_command}");
const r = exec("tmux", args, { encoding: "utf8", timeout: 5000 });
if (r.error) return null;
if (r.status !== 0) return false;
return panesRunPi(r.stdout ?? "");
}
// "Seen" marks: { "<project>/<agent>": "<lastActivity ISO>" }. A mark only
// applies while the agent's newest message still has that timestamp; anything
// the agent writes afterwards clears it automatically.
export function seenKey(rec) {
return `${rec.project}/${rec.agent}`;
}
export function seenPath(boardDir) {
return join(boardDir, "seen.json");
}
// Fail closed: a present but unreadable seen.json refuses the scan rather than
// silently dropping every mark.
export function loadSeen(boardDir) {
const path = seenPath(boardDir);
if (!existsSync(path)) return {};
let parsed;
try {
parsed = JSON.parse(readFileSync(path, "utf8"));
} catch (err) {
throw new ConfigError(`seen marks file is not valid JSON: ${path} (${err.message})`);
}
if (!parsed || typeof parsed !== "object" || Array.isArray(parsed)) throw new ConfigError(`seen marks file must be a JSON object: ${path}`);
for (const [k, v] of Object.entries(parsed)) {
if (typeof v !== "string") throw new ConfigError(`seen marks file has a non-string value for ${JSON.stringify(k)}: ${path}`);
}
return parsed;
}
export function saveSeen(boardDir, marks) {
mkdirSync(boardDir, { recursive: true, mode: 0o700 });
writeAtomic(seenPath(boardDir), marks);
}
// Set or clear one mark. Returns the updated map.
export function markSeen(boardDir, { project, agent, lastActivity, seen = true }) {
for (const [name, v] of Object.entries({ project, agent, lastActivity })) {
if (typeof v !== "string" || v.length === 0 || v.length > 512) throw new ConfigError(`${name} must be a non-empty string`);
}
if (project.includes("/")) throw new ConfigError("project must not contain '/'");
if (typeof seen !== "boolean") throw new ConfigError("seen must be true or false");
const marks = loadSeen(boardDir);
const key = seenKey({ project, agent });
if (seen) marks[key] = lastActivity;
else delete marks[key];
saveSeen(boardDir, marks);
return marks;
}
// One agent -> one status record.
export function scanAgent(spec, { isAlive = tmuxIsAlive, now = () => new Date(), seen = {} } = {}) {
const alive = isAlive(spec.tmux);
const file = findNewestSession(spec.sessionsDir);
const session = file ? readSession(file) : null;
const state = deriveState({ alive, session });
const scannedAt = now();
const lastActivity = session?.lastTimestamp ?? null;
const ageSeconds = lastActivity ? Math.max(0, Math.round((scannedAt.getTime() - Date.parse(lastActivity)) / 1000)) : null;
const needsYou = state === "waiting" || state === "error";
const isSeen = needsYou && lastActivity !== null && seen[seenKey(spec)] === lastActivity;
return {
agent: spec.agent,
project: spec.project,
state,
waitingOnYou: needsYou && !isSeen,
seen: isSeen,
alive,
tmux: spec.tmux,
sessionFile: file,
sessionId: session?.sessionId ?? null,
cwd: session?.cwd ?? null,
lastActivity,
ageSeconds,
lastAssistantText: session?.lastAssistantText ?? null,
lastError: session?.lastError ?? null,
skippedLines: session?.skippedLines ?? 0,
scannedAt: scannedAt.toISOString(),
};
}
// Repo agents: <repo>/.pi/state/<agent>/sessions, tmux default socket, session = agent.
export function discoverRepoAgents(repoRoot) {
const stateDir = join(repoRoot, ".pi", "state");
if (!existsSync(stateDir)) return [];
const project = basename(resolve(repoRoot));
return readdirSync(stateDir)
.filter((n) => existsSync(join(stateDir, n, "sessions")))
.sort()
.map((agent) => ({ agent, project, sessionsDir: join(stateDir, agent, "sessions"), tmux: { socket: null, session: agent } }));
}
// Fleet agents: <fleet>/<agent>/.pi/agent/sessions, tmux socket mosaic-fleet, session = agent.
export function discoverFleetAgents(fleetRoot, { project = "fleet", socket = "mosaic-fleet" } = {}) {
if (!existsSync(fleetRoot)) return [];
return readdirSync(fleetRoot)
.filter((n) => existsSync(join(fleetRoot, n, ".pi", "agent", "sessions")))
.sort()
.map((agent) => ({ agent, project, sessionsDir: join(fleetRoot, agent, ".pi", "agent", "sessions"), tmux: { socket, session: agent } }));
}
function writeAtomic(path, data) {
const tmp = `${path}.tmp-${process.pid}`;
writeFileSync(tmp, JSON.stringify(data, null, 2) + "\n", { mode: 0o600 });
renameSync(tmp, path);
}
// Scan every spec and write <boardDir>/sessions/<project>/<agent>.json plus index.json.
export function scan(specs, { boardDir, isAlive, now } = {}) {
if (!boardDir || !isAbsolute(boardDir)) throw new ConfigError("boardDir must be an absolute path");
const seen = loadSeen(boardDir);
const records = specs.map((spec) => scanAgent(spec, { isAlive, now, seen }));
for (const rec of records) {
const dir = join(boardDir, "sessions", rec.project);
mkdirSync(dir, { recursive: true, mode: 0o700 });
writeAtomic(join(dir, `${rec.agent}.json`), rec);
}
const generatedAt = (now ? now() : new Date()).toISOString();
const index = {
generatedAt,
counts: Object.fromEntries(STATES.map((s) => [s, records.filter((r) => r.state === s).length])),
waitingOnYou: records.filter((r) => r.waitingOnYou).map(seenKey),
seen: records.filter((r) => r.seen).map(seenKey),
sessions: records,
};
mkdirSync(boardDir, { recursive: true, mode: 0o700 });
writeAtomic(join(boardDir, "index.json"), index);
return index;
}