Afi

Afi - Claude Code Skill

Prerequisites: Install the CLI

This skill drives the afi-cli binary. You must verify the CLI is installed before invoking any command from this skill. If it is missing, install it first:

1. macOS / Linux:

   bashCopy code
   bash <(curl -fsSL https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/afi/install.sh)
   

2. Windows (PowerShell):

   powershellCopy code
   iwr -useb https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/afi/install.ps1 | iex
   

3. Verify: afi-cli --version
4. Ensure ~/.local/bin (macOS / Linux) or %LOCALAPPDATA%\Programs\msp-skills (Windows) is on $PATH.

The installer downloads the afi-cli and afi-mcp binaries into ~/.local/bin (macOS / Linux) or %LOCALAPPDATA%\Programs\msp-skills (Windows). It does not register the skill with your agent and writes no MCP client config - see mcp-install.md for that wire-up.

If --version reports "command not found" after install, the runtime cannot see the binary directory on $PATH. Do not proceed with skill commands until verification succeeds.

Afi has no other CLI, SDK, or MCP server - the only alternative is the web portal, and the vendor explicitly discourages API polling. This CLI walks the whole hierarchy into local SQLite once (fleet-sync), then answers fleet-wide questions (coverage-gaps, fleet-health, backup-stale, reconcile-licenses) offline. A guarded offboard command runs the vendor's own archive-then-release sequence with a verification gate the portal lacks.

When to Use This CLI

Use this CLI for any programmatic question about Afi SaaS backup fleets: backup coverage, nightly task health, quota breaches, archive staleness, license reconciliation, and the guarded archive-and-offboard sequence for departing employees. It is the right choice for MSP-scale questions across many tenants, where the portal forces a per-tenant walk and the rate-limited API punishes polling - sync once, then query the local store.

Anti-triggers

Do not use this CLI for:

• Performing restores, exports, or browsing actual backup data - the Afi public API does not expose restore/export initiation or data download; use the Afi web portal
• Creating or editing backup policies - policies are read-only in the public API; configure them in the portal
• Real-time monitoring or continuous polling of task status - Afi explicitly throttles and may suspend polling applications; use scheduled syncs instead
• Managing Afi portal user accounts, SSO, or notification settings - not part of the public API surface

Unique Capabilities

These capabilities aren't available in any other tool for this API.

Local state that compounds

• fleet-sync - Walk the whole Afi hierarchy - installations, orgs, tenants, then each tenant's resources, protections, policies, archives, quotas, and task stats - into local SQLite in one respectful pass.

  Run this first (and on a schedule): every fleet question afterward - coverage, staleness, licensing, scorecards - answers offline without tripping Afi's rate limits.

  bashCopy code
  afi-cli fleet-sync
  

• coverage-gaps - Find resources in a tenant that have no backup protection applied - the backup blind spots.

  Reach for this when asked whether every mailbox, site, or drive is actually being backed up - it answers fleet-coverage questions one API can't.

  bashCopy code
  afi-cli coverage-gaps --tenant 01F000000000000411Z1101G1Y --agent
  

• fleet-health - One table across all synced tenants: task success/failure counts, failed-task list, and quota breaches.

  Use this for the Monday-morning 'is the whole fleet green' sweep instead of walking tenants one portal tab at a time.

  bashCopy code
  afi-cli fleet-health --since 24h --agent
  

• resolve - Map a Microsoft 365 / Google Workspace ID, email, or name to the canonical Afi resource or tenant, including Multi-Geo fan-out.

  Run this before any offboard or audit to be certain which Afi object corresponds to an external-system identity.

  bashCopy code
  afi-cli resolve jane.doe@example.com --agent
  

• reconcile-licenses - Compare purchased subscription quantities against actually-protected resource counts per tenant to flag over- and under-provisioning.

  Use this for the weekly licensing drift check that today requires manual spreadsheet stitching across two portal screens.

  bashCopy code
  afi-cli reconcile-licenses --org 01F0ORG0000000000000000000 --json
  

• backup-stale - Flag protected resources whose most recent backup archive is older than a threshold - protected-but-silently-failing resources.

  Use this to catch the silent-failure class: resources that have a policy attached but whose backups quietly stopped landing.

  bashCopy code
  afi-cli backup-stale --max-age 48h --agent
  

• tenant-scorecard - One tenant's full backup posture: coverage percentage, last task status, newest/oldest archive age, and quota state.

  Generate this card when prepping a customer ticket or QBR - the per-tenant deep dive that complements fleet-health.

  bashCopy code
  afi-cli tenant-scorecard 01F000000000000411Z1101G1Y --agent --select tenant_id,coverage_pct,quotas_exceeded
  

Guarded workflows

• offboard - Safely back up a departing user's resource, verify the archive landed, then release the protection - refusing the irreversible step until the backup is confirmed.

  Use this when offboarding an employee so the final backup is verified before the M365/GWS seat is released - never lose a leaver's data.

  bashCopy code
  afi-cli offboard 01F0RESOURCE0000000000000A --tenant 01F0TENANT00000000000000B --policy 01F0POLICY000000000000000C --reason "employee departure"
  

Command Reference

applications - Manage applications

• afi-cli applications - Lists installations for the application identified by the authentication key.

orgs - Manage orgs

• afi-cli orgs create - Creates a new child organization.
• afi-cli orgs get - Retrieves an organization by its ID.

tenants - Manage tenants

• afi-cli tenants <id> - Retrieves a tenant by its ID.

Finding the right command

When you know what you want to do but not which command does it, ask the CLI directly:

bashCopy code
afi-cli which "<capability in your own words>"

which resolves a natural-language capability query to the best matching command from this CLI's curated feature index. Exit code 0 means at least one match; exit code 2 means no confident match - fall back to --help or use a narrower query.

Recipes

Monday fleet sweep

bashCopy code
afi-cli fleet-sync --skip-archives && afi-cli fleet-health --failed-only --agent

Refresh the fleet store (skipping the heavy archives walk), then list only the tenants with failed tasks or quota breaches.

Find backup blind spots in one tenant

bashCopy code
afi-cli coverage-gaps --tenant 01F000000000000411Z1101G1Y --agent --select items.name,items.kind,items.external_id

LEFT JOIN of resources against protections, narrowed to the three fields an agent needs to open tickets.

Offboard a departed employee safely

bashCopy code
afi-cli resolve jane.doe@example.com && afi-cli offboard 01F0RESOURCE0000000000000A --tenant 01F0TENANT00000000000000B --policy 01F0POLICY000000000000000C --reason "employee departure" --dry-run

Resolve the ambiguous email to the canonical Afi resource, then preview the guarded final-backup-verify-release sequence before running it for real.

Catch silently failing backups

bashCopy code
afi-cli backup-stale --max-age 48h --json

Protected resources whose newest archive is older than two days - the failure class no portal screen surfaces.

Licensing drift check

bashCopy code
afi-cli reconcile-licenses --org 01F0ORG0000000000000000000 --json

Purchased subscription quantities vs actually-protected resource counts, per tenant.

Auth Setup

Create an Application and API key in the Afi portal (org level: Configuration → Apps; tenant level: Service → Settings → Apps), then set AFI_API_KEY. The key is sent as the raw Authorization header value (Afi keys look like appkey-...; no Bearer prefix). Each Application supports two keys for rotation. Keys inherit the Application's installation scope - if a tenant or org is missing from results, check where the Application is installed.

Run afi-cli doctor to verify setup.

Agent Mode

Add --agent to any command. Expands to: --json --compact --no-input --no-color --yes.

• Pipeable - JSON on stdout, errors on stderr

• Filterable - --select keeps a subset of fields. Dotted paths descend into nested structures; arrays traverse element-wise. Critical for keeping context small on verbose APIs:

  bashCopy code
  afi-cli orgs get <org-id> --agent --select id,name,status
  

• Previewable - --dry-run shows the request without sending

• Offline-friendly - sync/search commands can use the local SQLite store when available

• Non-interactive - never prompts, every input is a flag

• Explicit retries - use --idempotent only when an already-existing create should count as success, and --ignore-missing only when a missing delete target should count as success

Response envelope

Commands that read from the local store or the API wrap output in a provenance envelope:

jsonCopy code
{
  "meta": {"source": "live" | "local", "synced_at": "...", "reason": "..."},
  "results": <data>
}

Parse .results for data and .meta.source to know whether it's live or local. A human-readable N results (live) summary is printed to stderr only when stdout is a terminal AND no machine-format flag (--json, --csv, --compact, --quiet, --plain, --select) is set - piped/agent consumers and explicit-format runs get pure JSON on stdout.

Agent Feedback

When you (or the agent) notice something off about this CLI, record it:

Copy code
afi-cli feedback "the --since flag is inclusive but docs say exclusive"
afi-cli feedback --stdin < notes.txt
afi-cli feedback list --json --limit 10

Entries are stored locally at ~/.local/share/afi-cli/feedback.jsonl. They are never POSTed unless AFI_FEEDBACK_ENDPOINT is set AND either --send is passed or AFI_FEEDBACK_AUTO_SEND=true. Default behavior is local-only.

Write what surprised you, not a bug report. Short, specific, one line: that is the part that compounds.

Output Delivery

Every command accepts --deliver <sink>. The output goes to the named sink in addition to (or instead of) stdout, so agents can route command results without hand-piping. Three sinks are supported:

Unknown schemes are refused with a structured error naming the supported set. Webhook failures return non-zero and log the URL + HTTP status on stderr.

Named Profiles

A profile is a saved set of flag values, reused across invocations. Use it when a scheduled agent calls the same command every run with the same configuration.

Copy code
afi-cli profile save briefing --json
afi-cli --profile briefing orgs get <org-id>
afi-cli profile list --json
afi-cli profile show briefing
afi-cli profile delete briefing --yes

Explicit flags always win over profile values; profile values win over defaults. agent-context lists all available profiles under available_profiles so introspecting agents discover them at runtime.

Exit Codes

Argument Parsing

Parse $ARGUMENTS:

1. Empty, help, or --help → show afi-cli --help output
2. Starts with install → ends with mcp → MCP installation; otherwise → see Prerequisites above
3. Anything else → Direct Use (execute as CLI command with --agent)

MCP Server Installation

1. Install the MCP binary (run the install script from the Prerequisites section, or see mcp-install.md for per-agent wire-up).
2. Register with Claude Code:

   bashCopy code
   claude mcp add afi-mcp -- afi-mcp
   

3. Verify: claude mcp list

Direct Use

1. Check if installed: which afi-cli If not found, offer to install (see Prerequisites at the top of this skill).
2. Match the user query to the best command from the Unique Capabilities and Command Reference above.
3. Execute with the --agent flag:

   bashCopy code
   afi-cli <command> [subcommand] [args] --agent
   

4. If ambiguous, drill into subcommand help: afi-cli <command> --help.