Agent skill

Tutorial Checker

by eclipse-ankaios in eclipse-ankaios/ankaios

Check Ankaios tutorials by executing safe documented shell steps and validating expected outputs.

Apache-2.0Auto-check passed

Install Tutorial Checker

skills CLI
$ npx skills add eclipse-ankaios/ankaios --skill tutorial-checker -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install eclipse-ankaios/ankaios tutorial-checker --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/eclipse-ankaios/ankaios.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.github/skills/tutorial-checker .claude/skills/tutorial-checker && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
tutorial-checker
GitHub stars
125
Token cost
~1.8k tokens
SKILL.md length
1,000 words
Files
1
Skills in repo
4
Repo updated
First seen
Licence
Apache-2.0

At a glance

Check Ankaios tutorials by executing safe documented shell steps and validating expected outputs.

  • Works in 6 steps: Parse markdown and extract shell code… → Split multiline shell blocks into… → Run reach step one by one. → …
  • SKILL.md covers Trigger behavior, Tutorial discovery, Temporary workspace (required) and Execution process (agent-driven), plus 4 more sections
  • Calls curl; reaches eclipse-ankaios.github.io

What it does

Tutorial Checker is an agent skill from eclipse-ankaios/ankaios. Check Ankaios tutorials by executing safe documented shell steps and validating expected outputs. Use this when asked to "check tutorials on main" or to verify local tutorial files.

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Eclipse Ankaios provides workload and container orchestration for embedded devices like automotive HPCs. The licence is Apache-2.0.

Example prompts

  • “check tutorials on main”
  • “/tutorial-checker”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Parse markdown and extract shell code blocks.
  2. Split multiline shell blocks into executable commands (respect line continuations using ).
  3. Run reach step one by one.
  4. For each command
  5. Validate expectations
  6. Validate human understandability

What it can do on your machine

Read from SKILL.md and the folder at commit ae46412. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • curl

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • eclipse-ankaios.github.io

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Tutorial Checker loads about 1.8k tokens when it runs. Until then it costs about 50 tokens; SKILL.md has 1,000 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~50
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from eclipse-ankaios/ankaios at commit ae46412, republished under its Apache-2.0 licence (© eclipse-ankaios). 1,000 words, ~1,833 tokens.

Download SKILL.mdSave it as .claude/skills/tutorial-checker/SKILL.md (or your agent's skills folder).
name
tutorial-checker
description
Check Ankaios tutorials by executing safe documented shell steps and validating expected outputs. Use this when asked to "check tutorials on main" or to verify local tutorial files.

Tutorial Checker Skill

Use this skill to validate tutorials by directly following tutorial steps as the agent (not by delegating logic to a large checker script).

Trigger behavior

  • If the user asks "check tutorials on main" (or equivalent), use main docs as default without asking for source.
  • If the user asks to check local tutorials, ask for the tutorial name (for example events for tutorial-events.md).

Tutorial discovery

  • Main tutorials:
    • Load https://eclipse-ankaios.github.io/ankaios/main/usage/installation/.
    • Discover all links containing /usage/tutorial- at runtime.
  • Local tutorials:
    • Discover files matching /workspaces/ankaios/doc/docs/usage/tutorial-*.md at runtime.
    • If user requested local mode without a name, ask for the name.

Temporary workspace (required)

  • If you need to create tutorial-generated files in the repository workspace, use a dedicated folder under target (for example target/tutorial-check-outputs) and clean it up at the end.
  • Create all required files and intermediate files inside the dedicated folder
  • Run file-dependent commands from that temp directory (or use absolute temp paths) so no accidental repo files are produced.
  • At the end, report the temp directory used and whether cleanup was performed.

Execution process (agent-driven)

For each selected tutorial:

  1. Parse markdown and extract shell code blocks.
  2. Split multiline shell blocks into executable commands (respect line continuations using \).
  3. Run reach step one by one.
  4. For each command:
    • Show progress before execution (Tutorial X/Y - Step A/B: <command>).
    • Classify step as execute, transform-and-execute, skip-manual, or skip-unsafe.
    • Systemd is not available inside the devcontainer, however if you use ankaios-start script that does not accept a startup manifest, just use ankaios-start and start on the manifest content dependent Ankaios agents additionally and apply the startup manifest of the tutorial with a simple ank apply instead. Classify the step as transform-and-execute and report the transformation in the final report.
    • Execute safe commands in terminal and capture stdout/stderr + exit code.
    • If a waiting time is required before the next command, don't combine it with the next command as this may affect allowed command detection and execution.
  5. Validate expectations:
    • If surrounding text states "should print" and a following text block exists, compare output to expected snippets.
    • Mark step as failed if command fails or expected snippets are missing.
    • Include a warning in the final report as this may indicate a problem with the tutorial or environment.
  6. Validate human understandability:
    • Review the tutorial text around each step and mark understandable or unclear.
    • Flag unclear items such as missing prerequisites, unexplained placeholders, ambiguous wording, missing expected result, or required manual context not stated.

Command details

  • The following commands are in the PATH:
    • ank, ank-agent, ankaios-start, ankaios-clean, podman, systemctl
  • If the checking is running in a container, systemctl commands for starting Ankaios shall be supplemented with ankaios-start and ank-agent. In case systemdctl was not tested, the user must be informed in the final report.
  • Disallow chained commands (for example cmd1; cmd2, cmd1 && cmd2, cmd1 || cmd2).
  • Transform streaming commands to non-blocking equivalents and execute the transformed command:
    • ank logs -f <workload> or ank logs --follow <workload> -> ank logs --tail 10 <workload>
    • mosquitto_sub ... -> bounded receive mode, e.g. add -C <count> and -W <seconds> so command exits automatically.
    • Report both original and transformed command in step output.
  • When server server IP substitution is required (for example containing <SERVER_IP>) use http://localhost:25551 and assume all execution is local.

Web form simulation (curl)

Some tutorials instruct the user to open a browser and interact with a web UI (e.g., enter a speed value). The agent can automate this with curl, but only under strict conditions:

Show full SKILL.md (427 more words)Show less
Activation (user confirms via chat UI)
  • When the agent encounters a web UI step, it executes the curl command in the terminal like any other command.
  • The VS Code chat UI will show the command to the user for approval before execution — the user accepts or rejects it directly in the UI.
  • No special flags or keywords are needed.
Scope restrictions (mandatory)

All of the following must be true before the agent issues any curl command:

  1. Localhost only — target must be 127.0.0.1 or localhost. Any other host is forbidden.
  2. Port from tutorial — the port number must appear in the tutorial text itself (e.g., http://127.0.0.1:5000). Never guess or scan for ports.
  3. Workload must be running — the workload serving the web UI must have been started by the agent during this tutorial run and confirmed as Running(Ok).
  4. Form discovery — before POSTing, fetch the page with curl -s <url> to extract the HTML form action and field names. Never invent field names.
  5. No external requests — curl must never target external URLs, APIs, or services outside the tutorial's locally running workloads.
Execution flow

When a tutorial step says something like "open browser at <url>" or "use the web UI to enter a value":

  1. Fetch the page HTML: curl -s <url>
  2. Parse the <form> element to determine: method, action path, and input field name attributes.
  3. Submit a reasonable test value: curl -s -X POST <url><action> -d "<field>=<value>"
  4. Verify the response contains a success indicator (e.g., confirmation text in HTML).
  5. If a consumer workload should receive the value, check its logs to confirm end-to-end delivery.
  6. Classify the step as transform-and-execute and report both the original instruction and the curl commands used.
Reporting

In the final report, list all curl commands executed under a separate "Web form simulation" subsection, including:

  • the original tutorial instruction
  • the curl discovery and submission commands
  • the response summary
  • whether end-to-end verification passed

Live run feedback

  • While running, continuously provide concise status updates:
    • current tutorial
    • current step
    • pass/fail/skip decision
    • short reason for skips/failures

Final report format

After all selected tutorials:

  1. Per tutorial: passed, failed, skipped counts.
  2. For each failure include:
    • exact command
    • output (or output tail if long)
    • expected snippets (if any)
    • likely cause / what may have gone wrong
  3. Human-understandability summary per tutorial:
    • only list unclear steps with their surrounding text and reason for being unclear
    • give a brief overall assessment of the tutorial's clarity and usability based on the number and severity of unclear steps
  4. End with: ask whether to try fixing detected problems.

© eclipse-ankaios, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .github/skills/tutorial-checker of eclipse-ankaios/ankaios.

Open the folder on GitHubat commit ae46412

Compare with similar skills

Tutorial Checker next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Tutorial Checker compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Tutorial Checker this skilleclipse-ankaios/ankaios125—~1.8kAutomated safety check: PassApache-2.0
Documentcodewhale-hq/Codewhale41k—~170Automated safety check: PassMIT
Safe Script ExecutionHKUDS/OpenSpace7.7k—~830Automated safety check: PassMIT
Document Generatorgarrytan/gstack136k—~11kAutomated safety check: NotesMIT
Document Gen Unicode SafeHKUDS/OpenSpace7.7k—~3kAutomated safety check: PassMIT
Document Gen Unicode SafeHKUDS/OpenSpace7.7k—~2.6kAutomated safety check: PassMIT

Similar skills

  • Document

    codewhale-hq/Codewhale

    Write or update repository/product documentation: README, user guides, API docs, architecture, migration notes, and changelog material.

    41k GitHub stars~170 tokensUpdated today
    DevelopmentAuto-check passed
  • Safe Script Execution

    HKUDS/OpenSpace

    Fallback workflow for code execution when automated tools fail, using shell heredoc with verification

    7.7k GitHub stars~830 tokensUpdated 1 mo ago
    Auto-check passed
  • Document Generator

    garrytan/gstack

    Writes missing documentation from scratch for a feature, a module or a whole project, organized as tutorial, how-to, reference and explanation pages.

    136k GitHub stars~11k tokensUpdated today
    DevelopmentAuto-check: notes
  • Unicode-safe fallback workflow for multi-format document generation with character sanitization

    7.7k GitHub stars~3k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed
  • Unicode-safe fallback workflow for multi-format document generation with character sanitization and explicit error capture

    7.7k GitHub stars~2.6k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed
  • Documents

    asgeirtj/system_prompts_leaks

    Help choose a written document's audience, purpose, structure, evidence, or destination.

    69k GitHub stars~1.2k tokensUpdated yesterday
    Auto-check passed

More from eclipse-ankaios/ankaios

  • Security Vulnerability Analysis

    eclipse-ankaios/ankaios

    Analyze potential Ankaios security vulnerabilities from pasted reports, local evidence, or advisory URLs.

    125 GitHub stars~1.5k tokensUpdated 2 days ago
    Auto-check passed
  • Examples Checker

    eclipse-ankaios/ankaios

    Check Ankaios examples by building and running each example in the devcontainer and validating they work correctly.

    125 GitHub stars~2k tokensUpdated 2 days ago
    Auto-check passed
  • Requirement Tracing

    eclipse-ankaios/ankaios

    Manage requirement tracing — write new or update existing requirements/design decisions, link implementations and tests to existing ones, and maintain tracing consistency.

    125 GitHub stars~2.4k tokensUpdated 2 days ago
    Auto-check passed

Questions about Tutorial Checker

What does Tutorial Checker do?

Check Ankaios tutorials by executing safe documented shell steps and validating expected outputs. Tutorial Checker is an agent skill from eclipse-ankaios/ankaios. Check Ankaios tutorials by executing safe documented shell steps and validating expected outputs.

How do I install Tutorial Checker in Claude Code?

Run `npx skills add eclipse-ankaios/ankaios --skill tutorial-checker -a claude-code`. Or copy the skill folder (.github/skills/tutorial-checker in eclipse-ankaios/ankaios) into .claude/skills/tutorial-checker in your project. Claude Code loads it when a task matches its description.

How do I install Tutorial Checker in Codex?

Run `npx skills add eclipse-ankaios/ankaios --skill tutorial-checker -a codex`. Or copy the skill folder (.github/skills/tutorial-checker in eclipse-ankaios/ankaios) into .agents/skills/tutorial-checker in your project. Codex loads it when a task matches its description.

Can I use Tutorial Checker in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add eclipse-ankaios/ankaios --skill tutorial-checker -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/tutorial-checker, .gemini/skills/tutorial-checker, .github/skills/tutorial-checker and .opencode/skills/tutorial-checker in your project.

What does Tutorial Checker need to run?

Going by SKILL.md and its folder, Tutorial Checker needs the command-line tools its instructions call (curl).

Does Tutorial Checker access the network?

SKILL.md names 1 domain. In commands or code: eclipse-ankaios.github.io; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Tutorial Checker safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Tutorial Checker use?

Tutorial Checker is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Tutorial Checker use?

About 1.8k tokens (SKILL.md is roughly 7.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Tutorial Checker?

Skills that share tags, products or a category with Tutorial Checker: Document (codewhale-hq/Codewhale, 41k stars), Safe Script Execution (HKUDS/OpenSpace, 7.7k stars), Document Generator (garrytan/gstack, 136k stars) and Document Gen Unicode Safe (HKUDS/OpenSpace, 7.7k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Tutorial Checker?

eclipse-ankaios (a GitHub organization) maintains it in eclipse-ankaios/ankaios, which has 125 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on October 5, 2026.

Source: eclipse-ankaios/ankaios on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.