SKILL.md- the skill, written out in plain English

   curl -XPOST 127.0.0.1:3147/chat-schedule \
     -H "Content-Type: application/json" \
     -d '{"intent":"morning brief","fireAt":"2026-05-01T08:00:00Z"}'

AGENTS.md- what the AI loads when this skill comes up

echo "[$(date -u +%FT%TZ)] chat-schedule: <what was missing or fixed> [FIXED|LOGGED] action_kind=skill-ran" >> state/log/loader-feedback.log

api.ts- the code it can call

#!/usr/bin/env npx tsx
/**
 * state/lib/chat-schedule.ts — create scheduled agents from the chat surface.
 *
 * Creates a new state/agents/<id>.json with {name, intent, cron} triplet.
 * Minted via snappy-chat POST /agent/create-scheduled.
 *
 * Schema: Agent from state/lib/agents.ts with optional schedule_cron.
 */

import { writeAgent, normalizeId } from "./agents.ts";
import type { Agent } from "./agents.ts";

export interface ScheduleInput {
  name: string;           // display name, kebab-case normalized
  intent: string;         // the prompt/directive the scheduler will run
  cron: string;           // 5-field cron expression (e.g. "0 9 * * 1")
  loader_slug?: string;   // optional explicit loader; defaults to inferred
}

export interface ScheduleOutput {
  ok: boolean;
  id?: string;
  name?: string;
  intent?: string;
  cron?: string;
  loader_slug?: string;
  error?: string;
}

/**
 * Validate cron expression: simple check for 5 space-separated fields.
 * Full validation deferred to scheduler.
 */
function isValidCron(cron: string): boolean {
  const parts = cron.trim().split(/\s+/);
  return parts.length === 5 && parts.every(p => p.length > 0);
}

/**
 * Create a scheduled agent from {name, intent, cron}.
 *
 * Returns {ok: true, id, name, intent, cron, loader_slug} on success.
 * Returns {ok: false, error: "..."} on failure.
 *
 * Idempotent: safe to call multiple times with same inputs (will re-write
 * the agent file but result is deterministic).
 */
export function createScheduledAgent(input: ScheduleInput): ScheduleOutput {
  // Validate inputs
  const name = (input.name || "").trim();
  if (!name || name.length === 0) {
    return { ok: false, error: "name required" };
  }
  if (name.length > 100) {
    return { ok: false, error: "name >100 chars" };
  }

  const intent = (input.intent || "").trim();
  if (!intent || intent.length === 0) {
    return { ok: false, error: "intent required" };
  }
  if (intent.length > 2000) {
    return { ok: false, error: "intent >2000 chars" };
  }

  const cron = (input.cron || "").trim();
  if (!cron || cron.length === 0) {
    return { ok: false, error: "cron required" };
  }
  if (!isValidCron(cron)) {
    return { ok: false, error: "cron must be 5 space-separated fields" };
  }

  const loader_slug = (input.loader_slug || "").trim();

  // Mint a collision-safe id: normalized name + short timestamp suffix
  const tsSuffix = Date.now().toString(36).slice(-6);
  const idBase = normalizeId(name).slice(0, 30) || "scheduled";
  const id = `${idBase}-${tsSuffix}`;

  // Build the agent record
  const agent: Agent = {
    id,
    status: "running",
    prompt: intent,
    ticks: 0,
    max_ticks: 1000,
    started_at: new Date().toISOString(),
    last_tick_at: null,
    last_tick_status: null,
    last_tick_dur_secs: null,
    loader_slug: loader_slug || undefined,
    schedule_cron: cron,
  };

  try {
    writeAgent(agent);
    return {
      ok: true,
      id,
      name,
      intent,
      cron,
      loader_slug: loader_slug || undefined,
    };
  } catch (e: any) {
    return { ok: false, error: e?.message || "writeAgent failed" };
  }
}

scripts- helper scripts it can run

prose-only skill - no sidecar under state/bin/ yet. Steps, if any, are described in SKILL.md.

how we check it- the checks, plus the last 3 runs