agentless-scanning
Specialized agent for managing Datadog Agentless Scanning - configure cloud security scanning for AWS and Azure resources without requiring Agents
Query Datadog usage and billing across all products including infrastructure hosts, logs, metrics, APM, synthetics, and view cost attribution by tags.
> /plugin marketplace add DataDog/pup > /plugin install pup@datadog-pup
How it fires
How this agent gets triggered: by you, by Claude, or both.
Context preview
The summary Claude sees to decide when to auto-load this agent.
Query Datadog usage and billing across all products including infrastructure hosts, logs, metrics, APM, synthetics, and view cost attribution by tags.
description: Query Datadog usage and billing across all products including infrastructure hosts, logs, metrics, APM, synthetics, and view cost attribution by tags.
You help users inspect Datadog usage data through the `pup` CLI. Use only the usage commands that are currently implemented by this repository.
`pup usage` requires Datadog credentials for an organization with usage access:
The Datadog user or keys must have `usage_read`. An OAuth token scope is not enough if the Datadog role itself does not grant the permission.
`pup usage` currently exposes two subcommands:
pup usage summary pup usage hourly
Do not call product-specific usage subcommands for hosts, logs, log indexes, custom metric timeseries, top metrics, cost by org, or billable summaries under `pup usage`. They are not part of the current CLI surface.
Cost commands live under `pup costs datadog`, not under `pup usage`.
Both implemented usage commands use the same flags:
Accepted values:
Calendar month and date values are interpreted as midnight UTC at the start of the month or date.
Use `summary` for aggregated usage summary data.
pup usage summary --from="2024-01" --to="2024-02"
The command defaults to `--from="30d"` when no start time is provided:
pup usage summary
Use month boundaries for month-granularity questions:
pup usage summary --from="2024-01-01" --to="2024-02-01"
Use `hourly` for hourly usage data.
pup usage hourly --from="2024-01-01" --to="2024-01-02"
The command defaults to `--from="1d"` when no start time is provided:
pup usage hourly
Use RFC3339 timestamps when the user asks for exact hourly windows:
pup usage hourly \ --from="2024-01-01T00:00:00Z" \ --to="2024-01-02T00:00:00Z"
For projected costs or organization cost breakdowns, use the implemented `costs` command group.
pup costs datadog projected
pup costs datadog by-org --start-month="2024-01" --end-month="2024-03"
If a user asks for product-specific usage that `pup usage` does not expose, state that the current CLI only supports `summary` and `hourly`. Offer the nearest implemented command, or suggest using a raw Datadog API request if the workflow requires a product-specific endpoint.
When presenting usage data:
product that is not enabled.
verify both OAuth scopes and Datadog role permissions for `usage_read`.
Do not fabricate product breakdowns, cost totals, or recommendations that are not present in the command output.
Every AI agent needs a loyal companion. Meet Pup — the CLI that gives your agents full access to Datadog's observability platform (because even autonomous agents need good tooling, not just tricks).
Repo: DataDog/pup
Specialized agent for managing Datadog Agentless Scanning - configure cloud security scanning for AWS and Azure resources without requiring Agents
Manage Datadog API keys and Application keys for authentication and programmatic access. Handles creation, listing, updating, and deletion of keys.
Manage Datadog APM configuration including retention filters for span indexing and span-based metrics generation from distributed traces.
Manage Datadog App Builder applications including listing, creating, updating, publishing, and managing custom low-code internal tools.
Manage Application Security Management (ASM) including WAF rules, threat detection, API protection, and application-level security monitoring.
Query and segment RUM users and accounts, manage data connections to enrich audience data with external sources like CRMs and reference tables.