Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add eliasstravik/gtm-skills --skill gtm-workspacegit clone --depth 1 https://github.com/eliasstravik/gtm-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/eliasstravik/gtm-skills/gtm-workspace)<a href="https://agentmods.dev/skills/eliasstravik/gtm-skills/gtm-workspace"><img src="https://agentmods.dev/badge/skills/eliasstravik/gtm-skills/gtm-workspace/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/eliasstravik/gtm-skills/gtm-workspace"><img src="https://agentmods.dev/badge/skills/eliasstravik/gtm-skills/gtm-workspace.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00157 | $0.01320 |
| Opus 5.5 | $0.00063 | $0.00528 |
| Sonnet 5.5 | $0.00031 | $0.00264 |
| Haiku 4.5 | $0.00016 | $0.00132 |
Grade A, and why
gtm-workspace scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 22d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GTM Workspace
Trigger
Apply this skill when a request concerns a GTM workspace as a whole, its organization facts, or its members.
Scope
One workspace holds one organization's GTM context as Markdown under ~/.gtm/<org-slug>/, with git as its memory. This skill owns the workspace root, ORG.md, members/, and workspace health, and hosts the shared standards in references/. ICPs, personas, fit checks, and workflows/ belong to the other gtm skills; Doctor never inspects the internals of workflows/.
Workflow hosting and provider connections belong to gtm-workflow. Route standalone Local or Vercel setup to its shared scripts/setup.mjs; do not require GTM Agent or Slack. Workspace Doctor delegates connection health to that skill's Doctor.
Inputs
The request; the workspace found by the discovery order in the contract; facts from the user, the company's own public site, and other safe public sources, never invented; a member's email from the user or a source, never inferred.
Roles
The user approves every change through the host's write permission; the agent proposes, edits, commits, and reports.
Procedure
The first gtm skill used in a conversation checks once for a newer release, as updates says. Talk by the six rules in interaction; reproduce the dialogues.
| Job | Do |
|---|---|
| Create | When ~/.gtm/ already holds an empty clone with a remote (empty, or holding only workflows/, .gitignore and .github/, which shared gtm-workflow setup publishes when it takes the workspace live), scaffold into it, take its directory name as the slug, do not ask where it lives, and after the commit run git push -u origin main; a clone that is neither empty nor a workspace (a GitHub-initialised README, say) is reported as a problem in business terms, never scaffolded over. Otherwise ask where it lives: this computer only (recommended), or also a private GitHub repository named gtm-<slug>, and git init on main. Copy templates/AGENTS.md, CLAUDE.md, ORG.md, and gitignore (as .gitignore) into ~/.gtm/<slug>/, fill ORG.md from the user and public sources, commit, and when shared run gh repo create gtm-<slug> --private --source . --push; the first push into an empty repository is git push -u origin main. Import is Create by copying or cloning an existing workspace into ~/.gtm/<slug>/ instead of scaffolding. |
| Update | Change ORG.md facts, or add, change, or remove members/<slug>/MEMBER.md from the template; email is required. |
| Doctor | Compare the root shape and every file to templates/ and the contract; report deviations in business terms; offer the fixes as options; rewrite what the user accepts, including legacy layouts (a root README, suborgs/, missing or misordered fields); .github/ belongs to gtm-workflow's Doctor. The workspace slug is the directory name, never derived from the H1. |
What ships with it
14 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/company-data.md 1.5 KB
- references/contract.md 2.8 KB
- references/interaction.md 3.7 KB
- references/interactions.md 4.2 KB
- references/person-data.md 966 B
- references/updates.md 5.2 KB
- scripts/check-update.mjs 5.0 KB runs code
- templates/AGENTS.md 1.2 KB
- templates/CLAUDE.md 11 B
- templates/gitignore 14 B
- templates/MEMBER.md 168 B
- templates/ORG.md 470 B
- tests/check-update.test.mjs 8.2 KB runs code
- tests/interaction.test.mjs 937 B runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 22d ago Changed e30e572d470b
- 23d ago Changed 812f9e5481f9
- 1mo ago Changed 01eba0680ae8
- 1mo ago Changed 830006fc2111
- 1mo ago Changed afdef924dcc7
- 1mo ago Changed · +2 lines 60ff17eb6c67
- 1mo ago Changed 350a37585a8b
- 1mo ago Changed 74e6addb84e3
- 1mo ago Changed · +2 lines 3d05bbb4172e
- 1mo ago First seen · 54 lines · 157 tokens per session scan A 1e2638efbbfa
gtm-workspace is a skill published in the GitHub repository eliasstravik/gtm-skills (24 stars, last pushed 22d ago), licensed MIT. It adds 157 tokens to every session and 1,320 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
gtm-workspace compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| gtm-workspaceeliasstravik/gtm-skills | Skill | 24 | 22d ago | MIT | A | 157 | Claude Code, Codex |
| startup-launch-playbookGingiris-1031/gingiris-skills | Skill | 89 | 2d ago | MIT | A | 58 | Claude Code, Codex |
| google-tag-managerVKirill/claude-lane-stack | Skill | 124 | 8d ago | MIT | A | 134 | Claude Code |
| content-atomizerSalesably/salesably-marketplace | Skill | 54 | 24d ago | MIT | A | 57 | Claude Code |
| orchestratorSalesably/salesably-marketplace | Skill | 54 | 24d ago | MIT | A | 49 | Claude Code |
| sales-orchestratorSalesably/salesably-marketplace | Skill | 54 | 24d ago | MIT | A | 46 | Claude Code |
| keyword-researchSalesably/salesably-marketplace | Skill | 54 | 24d ago | MIT | A | 54 | Claude Code |