# Audit Swan

> Checks whether a Swan workspace is structurally healthy. Use for configuration audits that surface missing knowledge, disconnected integrations, exhausted senders, broken triggers, sparse CRM data, and credit or subscription issues.

- **Type:** Skill
- **Install:** `agentstack add skill-swan-gtm-gtm-skills-audit-swan`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [swan-gtm](https://agentstack.voostack.com/s/swan-gtm)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [swan-gtm](https://github.com/swan-gtm)
- **Source:** https://github.com/swan-gtm/gtm-skills/tree/main/skills/ido-goldberg/audit-swan

## Install

```sh
agentstack add skill-swan-gtm-gtm-skills-audit-swan
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

## Instructions

### What this skill checks

Eight structural dimensions. Each is small — most produce 0-2 findings. Process them one at a time, write down the verdict (OK / WARN / FAIL) + a one-line summary, and move on. Do not try to gather all eight dimensions' raw data before reasoning. By the time you reach the report, you should have eight short verdicts in hand, nothing more.

---

### Dimension 1 — Subscription and credits

`swan-get-subscription`. Check three things:
- Plan tier (FREE / paid).
- Credit balance + recent burn rate. If you don't have burn data in the response, estimate from a small `swan-search-sequences` recent-activity sample (page 10) — don't pull the full log.
- Feature flags vs. what active plays use.

Verdict:
- FAIL: out of credits, or active plays use features not enabled.
- WARN:  7 days, or auth-broken channel that a play expects.
- OK: ≥ 1 healthy sender with capacity.

One-line summary.

---

### Dimension 4 — Integrations

For each integration referenced by an active trigger:
- Connected and authorized.
- Last successful sync recent.

If integration is referenced by an active trigger but not connected: FAIL.

Don't enumerate integrations the org doesn't use. One-line summary.

---

### Dimension 5 — Active triggers (gradual)

Pull active triggers, page size 10. For each:
- Required inputs available? (sender connected, list exists, integration present)
- Has it fired ≥ 1 time in 60 days, or is it new?

Most triggers are healthy. Stop after the first 10 unless ≥ 3 problems surface — then pull more.

Verdict per trigger group, summarized to one line.

---

### Dimension 6 — CRM hygiene

Only if a CRM is connected. Two tiny calls:
- `hubspot-list-object-properties` for companies. Check that the properties active plays depend on exist.
- `hubspot-search-objects` with page size 10 to sample populated-ness of `hubspot_owner_id` + lifecycle stage. 10 is enough to spot a hygiene problem.

Verdict:
- FAIL:  20 failures returned in the truncated preview: switch to `swan-execute-code`. Read `files/tool-outputs/`, group by `errorClass` in pandas, print classes with count > 3.

Verdict based on top class count: > 10 → FAIL, 3-10 → WARN, 

PAY ATTENTION:
  - 

ALL CLEAR:
  - 

NEXT STEP:
  
```

If no FAILs: lead with "no critical issues" before the WARN section.

---

### Rules

- MUST process dimensions sequentially. Each dimension's verdict + one-line summary is committed before the next runs.
- MUST cite specifics ("3 FAIL: 2 from HubSpot 401s, 1 from missing sender") not generalities ("integration issues").
- NEVER change anything. This skill is read-only.
- NEVER pull large logs in full. Use code + file-pointer if a list is > 20 items.
- If the org is < 7 days old, skip Dimensions 5-7 and label the report as "early-stage check — performance dimensions will populate as you use Swan."

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [swan-gtm](https://github.com/swan-gtm)
- **Source:** [swan-gtm/gtm-skills](https://github.com/swan-gtm/gtm-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-swan-gtm-gtm-skills-audit-swan
- Seller: https://agentstack.voostack.com/s/swan-gtm
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.