Trailblaze is a UI testing framework that lets coding agents control iOS, Android, and web devices through natural-language actions, then save those sessions as replayable test trails. It is for teams testing device-based user flows, with recorded actions that CI can replay without an LLM. The catalogue skills teach agents how to drive devices, save and replay trails, and create custom tools.
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 block/trailblaze --skill trailblaze-authorgit clone --depth 1 https://github.com/block/trailblazeWrote 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/block/trailblaze/trailblaze-author)<a href="https://agentmods.dev/skills/block/trailblaze/trailblaze-author"><img src="https://agentmods.dev/badge/skills/block/trailblaze/trailblaze-author/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/block/trailblaze/trailblaze-author"><img src="https://agentmods.dev/badge/skills/block/trailblaze/trailblaze-author.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Rogue Agent · line 137 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
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 | $0.00045 | $0.02436 |
| Opus 5.5 | $0.00018 | $0.00974 |
| Sonnet 5.5 | $0.00009 | $0.00487 |
| Haiku 4.5 | $0.00005 | $0.00244 |
Grade A, and why
trailblaze-author 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 yesterday.
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 — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Author a trail from a demonstration bundle
A human demonstrated a flow on a live device. Every interaction was captured with evidence. Your job is to produce a trail that runs on its own: deterministic where possible, resilient where the screen is dynamic, and proven by you actually running it before you call it ready. You are not transcribing clicks; you are authoring a test that validates what the human said they were validating.
Work in explicit phases, in order. Do not skip the audit passes and
do not claim ready without a passing verification run.
The demonstration bundle
The launching prompt gives you the bundle directory. A bundle holds
ONE platform's demonstration - bundles are keyed by platform (the
directory is named like demos/iphone/, demos/android/,
demos/android-tablet/), and sibling platform bundles from earlier
sessions may sit beside it. Inside:
| File | What it is |
|---|---|
demo.yaml |
Manifest: target, platform, device classifiers, the trailhead the human picked (name + args) or manual: true, and the human's stated objective + notes. |
actions.ndjson |
One JSON line per interaction, in order. phase: "setup" lines are how the human positioned the app before pressing Start; phase: "step" lines are the demonstrated flow itself. Each line carries the gesture (kind, coordinates or text), the hit-tested element, recorded tool YAML, ranked selector candidates, and evidence file names. |
start-state.png / start-state-hierarchy.txt |
The screen at the moment the human pressed Start. This is what the trailhead must reach. |
<seq>-before.png, <seq>-after.png, <seq>-*-hierarchy.txt |
Per-action evidence. The hierarchy text has one line per element: bounds, type, label, id, interactive flag. |
events/*.ndjson, network.ndjson |
Captured app event streams and network traffic, when available. Each line has timeMs; correlate to actions by time window (between one action's timeMs and the next). |
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.
- yesterday Changed · -5 tokens per session 0d3807f1608e
- 8d ago Changed · -4 lines · -51 tokens per session 91c42b35c6d7
- 15d ago Changed · +2 tokens per session 97074040ee7f
- 1mo ago First seen · 207 lines · 99 tokens per session scan A 8a4cf733fd02
trailblaze-author is a skill published in the GitHub repository block/trailblaze (322 stars, last pushed 2d ago), licensed Apache-2.0. It adds 45 tokens to every session and 2,436 once invoked, about $0.0002 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.
Other skills, from other repositories
android-ui-journey-testing
XML-specified Android UI journey testing, interactive step execution, assertion verification, and JSON outcome reporting.
android_ui_verification
Automated end-to-end UI testing and verification on an Android Emulator using ADB.
solopi-ai
A command-line framework for testing Android apps and devices with SoloPi, including on-device or cloud AI decision models. It manages devices, test cases, recorded interactions, replays, performance history, and evidence.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
eas-simulator
Run and control a user's app on a remote iOS/Android simulator hosted on EAS cloud. Read before running any eas simulator: commands - it has the current syntax for this experimental API. Use whenever the user needs a simulator they can't run locally - 'run my app on a cloud simulator', 'use eas simulator to…
test-native-extension
Validate a third-party control repo across four automated layers plus one printed manual recipe. Layer 1 asserts native-source structure (Android getName() and iOS +moduleName to manifest nativeModule; @ReactMethod / RCTEXPORTMETHOD to methods; no @ReactModule) plus load/init readiness (ReactPackage public no-arg…