list_categories
Play demoTry asking
- Which categories of logical fallacies are there?
- show the fallacy categories
ghost skill run ghost/logical-fallacies list_categories Logical Defence (F-Droid): list the logical fallacies, read one with its definition and example, group them by category. Offline, no account.
Agent run: finished without the skill, finished with it (not independently checked).
One recorded run each way, claude-haiku-5-5, same phone and same prompt. Watch it in the Before / after tab.
11.5× fewer tokens, 1.6× slower with the skill. Agent run: finished without the skill, finished with it (not independently checked).
| Run | Time | Tokens | Tool calls | Result |
|---|---|---|---|---|
| No skill: agent drives by hand | 32.7 s | 257,689 | 10 | finished |
| With skill: find_skill, run_skill | 50.7 s | 22,363 | 4 | finished |
Prompt: Open the Logical Defence app on my phone, list its categories of logical fallacies and explain the Appeal to Authority fallacy. Do not use the Play Store, no installs, no sign-in.
Same prompt, same phone, clean app state per side. Left: no skill, the agent drives the UI by hand. Right: find_skill + run_skill. 'ok' means the agent finished its turn; the result was not independently checked.
One recorded pair on samsung-a03s, claude-haiku-5-5; tokens count every input, cache and output token once per assistant message. Not a signed run: it adds no run evidence.
Control Logical Defence on Android with an AI agent. Look up logical fallacies on your phone: an agent lists them, reads definitions and examples and groups them by category.
Target search phrase: AI agent explains logical fallacies on Android
Demo shows: list_fallacies (category=On the Attack, limit=5)
Ghost in the Droid seed skill for Logical Defence (
za.co.lukestonehm.logicaldefence), an offline encyclopedia of logical fallacies. Every entry (number, name, definition, example) is readable text, so an agent can look fallacies up and explain them. Offline, no account.
| Action | What it does | Params |
|---|---|---|
list_categories | List the fallacy categories from the navigation drawer. | none |
list_fallacies | List the fallacies of a category (number and name); without a category the one on screen. | category, limit |
get_fallacy | Read one fallacy by its exact name from any category: number, category, definition and example. | name |
An agent can drive the whole app with these; compose them.
| Role | Action |
|---|---|
| Entry | list_categories |
| Primitives | list_fallacies |
| Read state | get_fallacy |
The app: F-Droid, or https://f-droid.org/packages/za.co.lukestonehm.logicaldefence/. The skill:
pip install "ghost-in-the-droid>=1.6.0" # 1.6.0 is not on PyPI yet: until it ships, git clone the repo and `pip install -e .`
ghost skill install ghost/logical-fallacies
ghost skill run ghost/logical-fallacies list_fallacies --param 'category=On the Attack' --device <serial>
ghost skill run ghost/logical-fallacies get_fallacy --param 'name=Appeal to Authority' --device <serial>
run_skill(name="ghost/logical-fallacies", action="get_fallacy", params={"name": "Appeal to Authority"})
# {"result": {"fallacy": {"number": 2, "definition": "...", "example": "..."}}}
Try asking: "List the categories of logical fallacies and explain Appeal to Authority."
Real Samsung phone, Android 13 (720x1600, 300 dpi, en-GB), Logical Defence 25. License: MIT.
| App version | Android | Device | Skill version | Dates |
|---|---|---|---|---|
| 25 | 13 | Samsung real phone (720x1600, 300 dpi, en-GB) | 1.0.0 | 2026-11-02 |
SOFT.md is the narrative an agent reads when the exact steps no longer match your app version and it has to adapt the skill.
What each action means, which UI landmarks it relies on and how success is recognised, so an agent can adapt it to another app of this kind or another version.
Package: za.co.lukestonehm.logicaldefence (Logical Defence, GPL-3.0, version 25 from the official F-Droid repository, sha256-verified)
Verified: 2026-11-02 on a real Samsung phone, Android 13, 720x1600 @ 300 dpi, en-GB. Offline, no account.
Skill class: HARD. No login, no checkpoints.
number / name / definition / example. The list scrolls.list_categories()Opens the drawer and reads the category entries, then closes it.
list_fallacies(category, limit)Selects the category in the drawer, scrolls the list from the top and collects the entry descriptions.
get_fallacy(name)Walks the categories until the entry shows and parses its description.
The skill reads entry descriptions, so another encyclopedia-style list app can reuse the same scroll-and-collect helper. get_fallacy needs the exact name as shown in list_fallacies.
Try a sample prompt or run an action by name. Replace any parameters in braces with your own values.
list_categoriesTry asking
ghost skill run ghost/logical-fallacies list_categories list_fallaciesTry asking
ghost skill run ghost/logical-fallacies list_fallacies get_fallacyTry asking
ghost skill run ghost/logical-fallacies get_fallacy Each release is an immutable git tag. The tree hash pins the exact skill directory the installer verifies.
| Version | Released | App versions | Tree | Size | Provenance |
|---|---|---|---|---|---|
1.0.0latest | never | >=1.0 | pending | 18.1 KB | untagged |