{"skill":{"slug":"parallel-monitor","displayName":"Parallel Monitor","summary":"Continuously track the web for changes via the Parallel Monitor API. Creates a recurring research task that runs on a cadence and emits events on change — wi...","description":"---\nname: parallel-monitor\ndescription: \"Continuously track the web for changes via the Parallel Monitor API. Creates a recurring research task that runs on a cadence and emits events on change — with optional webhook delivery. Use for 'monitor', 'track', 'watch', or 'alert me when' requests.\"\nhomepage: https://parallel.ai\n---\n\n# Parallel Monitor\n\nContinuously track the web for changes. A monitor is a recurring research task that runs on a cadence (hourly / daily / weekly / every two weeks) and emits an event each cycle. Optionally deliver events to a webhook.\n\n## When to Use\n\nTrigger this skill when the user asks for:\n- \"monitor [thing]\", \"track changes to [thing]\", \"watch [page/topic]\"\n- \"alert me when [X] changes / files / launches / posts\"\n- \"keep an eye on [pricing page / competitor / SEC filings]\"\n- Recurring research that should re-run automatically without re-prompting\n\n**Use Search for one-off lookups; use Monitor when the user wants ongoing observation.**\n\n## Quick Start\n\n```bash\nparallel-cli monitor create \"Track price changes for iPhone 16 Pro\" --json\n```\n\n## CLI Reference\n\n### Basic Usage\n\n```bash\nparallel-cli monitor create \"<objective>\" [options]\nparallel-cli monitor list --json\nparallel-cli monitor get mon_xxx --json\nparallel-cli monitor update mon_xxx [options] --json\nparallel-cli monitor delete mon_xxx\nparallel-cli monitor events mon_xxx --json\nparallel-cli monitor simulate mon_xxx --json\n```\n\n### Common Flags (create / update)\n\n| Flag | Description |\n|------|-------------|\n| `-c, --cadence` | `hourly`, `daily` (default), `weekly`, `every_two_weeks` |\n| `--webhook <url>` | Webhook URL for event delivery (required for `simulate`) |\n| `--output-schema <json>` | JSON schema for structured event output |\n| `--json` | Output as JSON |\n\n### Examples\n\n**Create a basic monitor (daily cadence):**\n```bash\nparallel-cli monitor create \"Track new SEC filings from Tesla\" --json\n```\n\n**Hourly monitor with webhook delivery:**\n```bash\nparallel-cli monitor create \"New AI funding announcements\" \\\n  --cadence hourly \\\n  --webhook https://example.com/hooks/funding \\\n  --json\n```\n\n**Structured-output monitor:**\n```bash\nparallel-cli monitor create \"Track iPhone 16 Pro pricing on apple.com\" \\\n  --output-schema '{\"properties\":{\"price_usd\":{\"type\":\"number\"},\"in_stock\":{\"type\":\"boolean\"}}}' \\\n  --json\n```\n\n**List existing monitors (default 10):**\n```bash\nparallel-cli monitor list --json\n```\n\n**Inspect events from a monitor:**\n```bash\nparallel-cli monitor events mon_xxx --json\n```\n\n**Test webhook delivery without waiting for a real cycle:**\n```bash\nparallel-cli monitor simulate mon_xxx --json\n```\n\n(`simulate` requires a `--webhook` was set on the monitor; otherwise it errors.)\n\n**Update cadence on an existing monitor:**\n```bash\nparallel-cli monitor update mon_xxx --cadence weekly --json\n```\n\n**Delete a monitor:**\n```bash\nparallel-cli monitor delete mon_xxx\n```\n\n## Best-Practice Prompting\n\n### Objective\nWrite 1-3 sentences describing:\n- What signal the monitor should look for (e.g., \"price changes\", \"new filings\", \"site outage\")\n- What \"change\" means (numeric threshold, new entity appearing, status flip)\n- Source preference if relevant (e.g., \"from sec.gov\", \"on the official pricing page\")\n\n### Cadence\nPick the slowest cadence that still catches the signal in time:\n- `hourly` — breaking news, pricing fluctuations, status pages\n- `daily` (default) — most ongoing tracking\n- `weekly` / `every_two_weeks` — slow-moving topics, regulatory filings, executive changes\n\n### Output Schema (optional)\nProvide a JSON schema when you want structured per-event output (e.g., `{price_usd, in_stock}`). The Monitor API will populate that schema each cycle, making downstream automation easier.\n\n## Response Format\n\n`monitor create` and `monitor list` return JSON with fields like:\n- `monitor_id` — `mon_xxx`\n- `objective` — the natural-language description\n- `cadence` — current schedule\n- `webhook` — delivery URL (if set)\n- `created_at` / `updated_at` — timestamps\n- `last_event_at` — most recent event timestamp (if any)\n\n`monitor events` returns an array of `{event_id, observed_at, output}` objects.\n\n## Output Handling\n\nWhen relaying monitor results to the user:\n- Always echo the `monitor_id` so the user can manage the monitor later\n- Default `monitor list` to `-n 10` to avoid megabyte-scale dumps in long-lived workspaces\n- For `events` output, summarize the trend across cycles, not just the latest event\n\n## Running Out of Context?\n\nFor long-running monitor management sessions, save state and use `sessions_spawn`:\n\n```bash\nparallel-cli monitor list --json -o /tmp/monitors.json\n```\n\nThen spawn a sub-agent:\n```json\n{\n  \"tool\": \"sessions_spawn\",\n  \"task\": \"Read /tmp/monitors.json and produce a summary of active monitors grouped by cadence.\",\n  \"label\": \"monitor-summary\"\n}\n```\n\n## Error Handling\n\n| Exit Code | Meaning |\n|-----------|---------|\n| 0 | Success |\n| 1 | Unexpected error (network, parse) |\n| 2 | Invalid arguments |\n| 3 | API error (non-2xx) |\n\n## Prerequisites\n\nRequires `parallel-cli` (installed and authenticated). If `parallel-cli --version` fails, or if a later command fails with an authentication error, tell the user to see https://docs.parallel.ai/integrations/cli and stop.\n\n## References\n\n- [API Docs](https://docs.parallel.ai)\n- [Monitor API Reference](https://docs.parallel.ai/api-reference/monitor)\n","tags":{"latest":"1.0.3"},"stats":{"comments":0,"downloads":349,"installsAllTime":13,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1778181610972,"updatedAt":1778492872010},"latestVersion":{"version":"1.0.3","createdAt":1778181610972,"changelog":"parallel-monitor 1.0.3\n\n- Added a detailed SKILL.md for usage, prompting, CLI examples, error codes, and best practices.\n- Clarified when to use monitoring versus one-off searches.\n- Outlined support for structured outputs via JSON schema and webhook event delivery.\n- Documented error handling procedures and prerequisites.\n- Included references to official documentation for further information.","license":"MIT-0"},"metadata":{"setup":[],"os":null,"systems":null},"owner":{"handle":"normallygaussian","userId":"s173n6zpcxryc8pb92y0hgm1f985dyrd","displayName":"NormallyGaussian","image":"https://avatars.githubusercontent.com/u/3804139?v=4"},"moderation":null}