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 tsouth89/toolport --skill toolportgit clone --depth 1 https://github.com/tsouth89/toolportWrote 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/tsouth89/toolport/toolport)<a href="https://agentmods.dev/skills/tsouth89/toolport/toolport"><img src="https://agentmods.dev/badge/skills/tsouth89/toolport/toolport/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/tsouth89/toolport/toolport"><img src="https://agentmods.dev/badge/skills/tsouth89/toolport/toolport.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00000 | $0.00802 |
| Opus 5 | $0.00000 | $0.00401 |
| Sonnet 5 | $0.00000 | $0.00160 |
| Haiku 4.5 | $0.00000 | $0.00080 |
Grade A, and why
toolport 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 9d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Working through Toolport
Toolport is a local gateway that aggregates every MCP server the user has set up. Instead of hundreds of tool definitions, you see a few meta-tools and discover the real tools on demand.
Core workflow
- Search first. For any external action, call
toolport_search_toolswith keywords describing the capability ("list emails","create payment","recent deployments"). If the service is connected, its tool is here, so do not tell the user a capability is unavailable until you have searched. - Call it. The first matching result includes its exact name and full input
schema and is ready to use: call
toolport_call_toolwith thatnameand all parameters inside theargumentsobject. Don't keep searching for a better match, and never invent identifiers (teamId, projectId, and so on). Fetch them with a list/get tool on the same server first. - Orient when needed.
toolport_statuslists every connected server, its tool count, and the tokens Toolport has saved. Passservertotoolport_search_tools(with an emptyquery) to list one server's full tool set.
Search tips
- Tool names are namespaced per server:
stripe__create_refund. - If the result says more tools matched than were shown, narrow with
serveror raiselimitbefore concluding anything is missing. - Many servers expose a generic API bridge (one write/create tool), so search by capability, not exact operation names.
Multi-step work: toolport_run_script
When you already know the steps, run ONE JavaScript orchestration script
server-side instead of many round-trips: servers.stripe.create_refund({...})
(sync) or .async({...}) with Promise.all to fan out. Intermediate results
stay full-sized inside the script; only your returned value is shaped for
context. Pass validate: true for a dry run that compiles the script and
returns the plan without executing. If a script fails partway,
structuredContent.toolportScript.progress lists which calls already ran
(their side effects are committed), so resume by index, not tool name.
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.
- 9d ago First seen · 62 lines · 0 tokens per session scan A d09ff6c72db9
toolport is a skill published in the GitHub repository tsouth89/toolport (202 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 802 tokens. 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.
Other skills, from other repositories
maestro
Direct Maestro command hub for Codex slash menu: compose or choose a Frontier model panel, settings, terse, and update.
terse
Token-efficient terse narration mode. Cuts agent explanation tokens while preserving requested artifacts. Levels: lite, full, ultra. Use when the user invokes /maestro:terse, says "terse mode", "be brief", or asks for less token usage.
frontier
Maestro Frontier local multi-CLI fusion engine — switch mode, or run a prompt through the panel.
maestro-settings
View and change Maestro toggles (terse, frontier, context-bar, discipline, verify) via the settings CLI.
maestro-terse
Toggle Maestro terse narration level (lite, full, ultra, off) via the settings CLI.
maestro-update
Update Maestro to the latest marketplace version for Codex.