agentleFS
BenPremBenPrem/business-brain/.claude/skills/access-transfer-tracker/SKILL.md

access-transfer-tracker

Skill1 starsChanged 3 months agoMITView on GitHub

Per-client access/asset transfer matrix for agency handoffs — asset, owner, transfer method, status — with verified-only-after-a-real-login discipline, a chase list with drafted nudges, and <TASK SYSTEM> sync. Triggers - "access matrix", "what access are we missing", "transfer status", "account handoff", taking over a client from another agency. NOT for agent tool connections — use mcp-integration.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Hard gate: never change a single setting on an asset still under the losing agency's control until ownership is verified — verification is …

    line 63

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Access Transfer Tracker

Tracks every access and asset handoff when you take over a client from another agency (or from the client's own scattered accounts). One living matrix per client; nothing counts as "have it" until proven by a real authenticated action.

The matrix file

Location: clients/<slug>/access/access-matrix.md (mkdir -p the access/ dir on first run). Frontmatter: type: access_matrix, client: <slug>, updated: YYYY-MM-DD. If the engagement started with a client access-intake note, keep that note as a dated record — never rewrite it; the living matrix supersedes it.

Standard asset rows (add/remove per client; for multi-business clients, one status column per business line):

AssetWhat "admin-level" means here
Domain registrarAccount access or a transfer-out auth code
DNS zone controlCan view/export AND edit records (often differs from registrar)
Hosting / CMSAdministrator role, not Editor
Google Business ProfileManager/Owner invite to your Google account
Social ads (Meta or equivalent)Page + ad account + pixel shared to YOUR business manager as a partner
Search ads / LSAAdmin invite or manager-account link
Analytics / tag managerAdministrator on the property/container
Search ConsoleDelegated Owner (not just Full user)
Email / workspace adminAdmin-console access or a documented sending setup
Review platformsBusiness-profile access covered above; other platform logins listed separately
Social account loginsNote the 2FA plan — 2FA tied to a client's personal phone is a standing friction to record
<CRM> / call trackingCRM admin login + ownership of any tracking numbers

Fields per row: current owner (client / losing agency / unknown) · access level needed · transfer method (invite to your account, credential handover, registrar transfer, partner share) · status: not-requested → requested → received → verified · verified date + how.

Credentials never go in the matrix or any workspace markdown. Secrets live in a gitignored env/credentials file — verify that file actually exists before directing anyone to it. The matrix stores only who has access, never the secrets themselves.

Verification discipline — received is not done

Status reaches verified ONLY after a real authenticated action, recorded with date + method:

  • Hosting/CMS: log in as the granted admin, or make an authenticated API call that returns real site data
  • Registrar/DNS: view or export the zone from inside the account; prove edit rights by reading a record you could change — don't change it (see the hard gate below)
  • Business profile: the location appears in YOUR profile manager with the granted role
  • Social ads: the page/ad account/pixel shows under YOUR business manager with the expected role
  • Analytics/Search Console: a UI or API read returns the client's real property data
  • Workspace: the admin console loads under your access

Credential handovers: before testing, grep any saved config for placeholder strings (PASTE_, YOUR_, xxx) — placeholders get pasted literally and then reported as "received."

Hard gate: never change a single setting on an asset still under the losing agency's control until ownership is verified — verification is read-only. DNS changes, business- profile edits, and live-site publishes each need the founder's or the client's explicit yes regardless of access status.

Outputs each run

  1. Updated matrix — bump updated; adjust statuses only with evidence (a forwarded invite = received; your own successful login = verified).
  2. Chase list — a section at the bottom of the matrix: every non-verified row → who to nudge (client vs. losing agency vs. platform support), exactly what to ask for, and which deliverable it blocks. Consolidate: ONE bundled request per recipient beats a drip of asks. Draft the actual nudge lines to clients/<slug>/_drafts/YYYY-MM-DD/access-nudges.md for the founder's review — never send directly. Voice: client = plain and direct, no agency-speak; losing agency = professional "we", factual, zero blame. Never state platform facts you haven't verified.
  3. <TASK SYSTEM> sync — one task per outstanding transfer (created at requested), a comment on every status change, completed only at verified. Confirm each API call returned a real payload before claiming the sync happened — never report task-system writes on exit code alone.

New-client takeover

mkdir -p clients/<slug>/access/, copy the standard rows, all statuses not-requested. The first deliverable is the consolidated request: one email to the client (what they own), one to the losing agency (what they hold). Route the drafting through email-composer — this skill produces the content requirements, not the relationship email.

Write-back (every run that changes anything)

  • Log an activity note on the client's workspace record (append-only — never modify existing activity records)
  • Mirror status changes into the <TASK SYSTEM> tasks the same session

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.