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 agentmods add skills/jazelly/tokenless/tokenless-installnpx skills add jazelly/tokenless --skill tokenless-installgit clone --depth 1 https://github.com/jazelly/tokenlessWhat 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 | $0.00056 | $0.01383 |
| Opus 5 | $0.00028 | $0.00691 |
| Sonnet 5 | $0.00011 | $0.00277 |
| Haiku 4.5 | $0.00006 | $0.00138 |
Grade A, and why
tokenless-install 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 2d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tokenless installation and maintenance
Use this workflow only for the maintenance task the user explicitly requested; it is not a prerequisite for every provider job. Execute the documented noninteractive maintenance and verification commands yourself. Managed-profile initialization and import are user-run workflows outside the agent session: never initiate either one or traverse providers on the user's behalf. Report status, manual actions, and final results in the user's preferred language, inferred from the conversation.
Install
-
Require Node.js 22.13 or newer:
node --versionIf it is missing or older, report the exact requirement and stop until the user installs it.
-
Install the latest CLI:
npm install --global tokenless@latestAfter installation, invoke only the global
tokenlesscommand. Never run the Tokenless CLI throughnpx tokenlessornpx tokenless@latest. -
Run the canonical maintenance workflow to reconcile the global skills and matching local daemon:
tokenless upgrade --json
upgrade already runs the final read-only doctor --json check. Report full success only when its top-level result is ok: true and its npmInstall, resolveGlobalCli, skills, runtimeInstall, and doctor phases are healthy. Summarize skill, browser, managed profile, daemon, worker, and provider readiness without exposing account identity or authentication data. If the CLI, skills, and local runtime phases succeed but doctor reports that no managed profile exists, report that the CLI installation completed and provider readiness is pending user-run profile initialization. Do not start that workflow for the user.
User handoff
Run the CLI step first. Pause only for a user-only browser action such as signing in, CAPTCHA, plan or permission UI, or provider confirmation.
Report the handoff in the user's preferred language with exactly three short parts:
- Completed locally: state which installation steps succeeded.
- Action needed: give the exact visible action and name the managed profile and provider when known.
- Next verification: ask the user to reply when finished, then run
tokenless profiles status --profile <slug> --provider <id> --jsonandtokenless doctor --json.
What ships with it
1 file 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.
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.
- 2d ago First seen · 98 lines · 56 tokens per session scan A 7202873f32fe
tokenless-install is a skill published in the GitHub repository jazelly/tokenless (11 stars, last pushed 3d ago), licensed MIT. It adds 56 tokens to every session and 1,383 once invoked, about $0.0003 per session on Opus 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.
Other skills, from other repositories
anakinscraper
Scrape any website into clean markdown or structured JSON. Anti-detect browser, smart proxy rotation, AI-powered data extraction.
pawflow-helper
Guide a user through the PawFlow interface with the non-destructive pawflow-ui semantic tool and optional avatar gestures.
setup
Bootstrap the Autoframe framework for a specific product: asks about the app, then wires a ready-to-run Playwright suite. Use when the user runs /setup, asks to set up or initialize Autoframe, or wants to start a test suite for their app.
browser
Use when automating browser interactions, web scraping, or E2E UI testing using AI-optimized abbreviated selectors and isolated multi-session contexts.
claude-in-chrome
Use when automating web browsing, page interactions, or DevTools via Chrome browser.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.