---
name: hive-stateful-monitoring
description: "Use this skill when a user asks Hive to remember, monitor, schedule, alert, or report on crypto intelligence across sessions. Creates durable monitor intent through Hive's stateful monitoring tools."
license: MIT
metadata:
  package: "@hiveintelligence/agent-skills"
  category: "stateful-monitoring"
  requires_network: "true"
version: 1.0.0
---

# Stateful Monitoring

Create, list, update, and archive durable crypto intelligence monitors that agents can schedule, remember, and report from.

## Trigger

Use this skill when a user asks Hive to remember, monitor, schedule, alert, or report on crypto intelligence across sessions. Creates durable monitor intent through Hive's stateful monitoring tools.

## Use when

- remember this
- monitor wallet
- watch token
- schedule crypto report
- alert me
- persistent crypto memory

## Source

Canonical source path: `agent-skills/packs/hive-stateful-monitoring/SKILL.md`

## Task toolsets

- `stateful_monitoring`: Create and manage durable crypto intelligence monitors that remember user intent across agent sessions. Required identifiers: monitor kind, target object.

## Procedure

1. Convert the user's durable intent into a monitor `kind`, `target`, `rules`, and `cadence` before writing state.
2. Use `hive_create_monitor` for new watch requests and `hive_list_monitors` before creating duplicates or changing an existing monitor.
3. Use `hive_update_monitor` to change cadence, rules, target, metadata, or status; use `hive_archive_monitor` when the user asks Hive to stop watching.
4. Use `hive_get_monitor_runs`, `hive_list_observations`, `hive_list_alerts`, `hive_get_latest_snapshot`, or `hive_generate_monitor_report` when the user asks what changed or what Hive remembers.
5. Use `hive_update_alert_status` for acknowledged, resolved, or reopened alerts.
6. Use `hive_remember_fact`, `hive_list_memory_facts`, and `hive_forget_memory_fact` for durable user-scoped memory that should appear in future monitor reports.
7. Use `hive_list_subjects`, `hive_get_subject`, `hive_archive_subject`, and `hive_list_subject_audit_events` only for B2B adapter state-boundary operations.
8. Explain that monitor execution is handled by Hive workers, not by the current chat session.

## Notes

- Worker-supported monitor kinds are `wallet`, `token`, `protocol`, `market`, `prediction_market`, `watchlist_digest`, `token_discovery_risk`, and `risk_watch`.
- For B2B adapter mode, state is isolated by server-injected subject context. Never ask the model or end user to provide `__hive_user_id` or `__hive_subject_id`.
- If stateful tools return `missing_key` or an auth-required error, say the monitor was not saved and explain which backend or auth prerequisite is missing.

## Runtime status handling

Hive uses `ok`, `missing_key`, `plan_required`, `rate_limited`, `degraded`, `failing`.
A non-`ok` state means the tool exists but the runtime cannot complete that call
right now. Do not hide or remove the tool.

## Source of truth

The executable MCP skill pack ships with the Hive MCP package. This public file
is a website mirror for agent discovery. Runtime metadata is available at
`https://mcp.hiveintelligence.xyz/mcp` via `hive://skills`. Installable skills are mirrored at
https://github.com/hive-intel/hive-skills; install the full pack with
`npx skills add hive-intel/hive-skills`.