How arvind3/robot-framework-py-skill works
1. Repository purpose
- This is not an application runtime; it is a content-and-validation repository for a Robot Framework skill.
- Its main job is to define authoritative guidance for generating, reviewing, and refactoring Robot Framework automation across five libraries: SeleniumLibrary, RequestsLibrary, Browser/Playwright, AppiumLibrary, and DatabaseLibrary.
2. Canonical skill content
- SKILL.md is the primary authored skill document.
- It contains the golden patterns, setup guidance, pitfalls, and do/don’t rules that the skill is expected to enforce.
- CHANGELOG.md records evolution of the skill.
3. Policy and acceptance criteria
- The repository treats references/acceptance-criteria.md as a governance artifact.
- The same acceptance criteria are mirrored under .github/skills/robot-framework-py/references/acceptance-criteria.md, so the skill packaging and the repo-local references stay aligned.
- These files define what “correct” skill behavior means and are used as the basis for validation.
4. Example suite subsystem
- examples/ is the practical demonstration layer.
- Each library gets its own directory with:
- a smoke Robot suite,
- an enterprise Robot suite,
- reusable resources,
- and environment-specific variables/env.py files.
- This gives the repo a pattern library for resource-driven Robot automation rather than a single monolithic example.
5. Library-specific boundaries
- The examples are cleanly separated by automation domain:
- examples/selenium/
- examples/browser/
- examples/requests/
- examples/appium/
- examples/database/
- Each domain has its own resource files and local test fixtures, which keeps UI, API, mobile, and database concerns isolated from each other.
6. Validation and quality subsystem
- tests/quality/test_skill_quality.py is the main unit-style quality gate.
- It enforces repository policy such as presence/synchronization of acceptance criteria and required skill artifacts.
- tests/scenarios/robot-framework-py/scenarios.yaml defines scenario-driven checks that validate the skill behavior across planned cases, including basic usage, error handling, advanced features, and CI-friendly mock flows.
7. End-to-end verification flow
- The expected validation path is:
- update skill content and criteria,
- run the quality tests,
- run the scenario harness with mocks when needed,
- then review PR evidence.
- This means the repo’s “data flow” is mostly documentation and validation flow, not runtime app data flow.
8. CI and pull-request gating
- .github/workflows/ci.yml is the main automation entry point for checks.
- .github/workflows/validate-skill.yml supports skill validation.
- .github/pull_request_template.md forces contributors to provide evidence for the required checks, making the quality process part of the repo’s architecture rather than an informal convention.
9. Published documentation site
- docs/ contains a static site (index.html, engineering-blog.html, site.js, styles.css).
- .github/workflows/deploy-pages.yml publishes this documentation to GitHub Pages.
- The docs site is an output surface for the skill, distinct from the skill source and test gates.
10. Operational/configuration artifacts
- agents/openai.yaml and analytics.config.json are supporting configuration files rather than core logic.
- They indicate the repo also expects agent/tooling integration and site analytics, but these are secondary to the skill content and validation pipeline.
11. External runtime dependencies
- The examples are designed to be executed with the real Robot Framework ecosystem: Robot Framework itself plus the relevant library packages.
- The Browser examples additionally depend on Playwright initialization (rfbrowser init), which matters because it changes the runtime setup boundary for that library.
12. Architecture summary for a diagram
- The cleanest diagram would show three main layers:
- Authoritative skill content and policy (SKILL.md, acceptance criteria, changelog),
- Executable examples and scenario/quality validation (examples/, tests/),
- Delivery and publication (.github/workflows/, docs/ GitHub Pages).
- The key cross-cutting flow is: skill guidance → examples/scenarios → automated checks → published docs / PR gate.
Main entry points: SKILL.md SKILL.md · health-check.robot Health check · index.html Docs site · acceptance-criteria.md Criteria
Components 20
Skill content
Authoritative guidance, policy, and release history for the Robot Framework skill.
- SKILL.md · Primary skill document with patterns, setup guidance, and do/don't rules.
SKILL.md - Criteria · Repo-local acceptance criteria that define correct skill behavior.
references/acceptance-criteria.md - Packaged criteria · Mirrored acceptance criteria used by the packaged skill reference.
.github/skills/robot-framework-py/references/acceptance-criteria.md - CHANGELOG · Evolution of the skill content and policy over time.
CHANGELOG.md
Examples
Executable Robot Framework examples separated by automation domain.
- Selenium · SeleniumLibrary example suites and shared resources.
examples/selenium - Browser · Browser/Playwright example suites and setup.
examples/browser - Requests · RequestsLibrary example suites and resources.
examples/requests - Appium · AppiumLibrary example suites and mobile resources.
examples/appium - Database · DatabaseLibrary example suites and resources.
examples/database - Health check · Repository-wide smoke-style Robot check used as a simple cross-cutting example.
examples/health-check.robot
Validation
Quality gates, scenario checks, and contributor evidence requirements.
- Quality gate · Main repository policy test for required artifacts and synchronization.
tests/quality/test_skill_quality.py - Scenarios · Scenario-driven checks covering usage, errors, advanced features, and mock flows.
tests/scenarios/robot-framework-py/scenarios.yaml
Delivery
CI, validation workflows, and published documentation site.
- PR template · Pull request checklist that requires validation evidence from contributors.
.github/pull_request_template.md - CI · Primary continuous integration entry point for checks.
.github/workflows/ci.yml - Validate skill · Dedicated workflow for skill validation runs.
.github/workflows/validate-skill.yml - Deploy pages · Workflow that publishes the docs site to GitHub Pages.
.github/workflows/deploy-pages.yml - Docs site · Published documentation surface for the skill.
docs/index.html - Blog page · Secondary documentation page within the published site.
docs/engineering-blog.html
Other components
- Agent config · Supporting configuration for agent/tool integration.
agents/openai.yaml - Analytics · Supporting analytics configuration for the docs or site tooling.
analytics.config.json
Connections 25
- SKILL.md → Criteria: defines · Skill guidance is constrained by acceptance criteria.
- Criteria → Packaged criteria: mirrored by · Repo criteria are kept aligned with packaged skill criteria.
- SKILL.md → CHANGELOG: evolves · Skill changes are tracked over time.
- SKILL.md → Selenium: guides · Authoritative patterns are demonstrated in Selenium examples.
- SKILL.md → Browser: guides · Authoritative patterns are demonstrated in Browser examples.
- SKILL.md → Requests: guides · Authoritative patterns are demonstrated in Requests examples.
- SKILL.md → Appium: guides · Authoritative patterns are demonstrated in Appium examples.
- SKILL.md → Database: guides · Authoritative patterns are demonstrated in Database examples.
- Selenium → Health check: contributes to · UI examples feed the shared smoke-style check.
- Browser → Health check: contributes to · Browser examples feed the shared smoke-style check.
- Requests → Health check: contributes to · API examples feed the shared smoke-style check.
- Appium → Health check: contributes to · Mobile examples feed the shared smoke-style check.
- Database → Health check: contributes to · Database examples feed the shared smoke-style check.
- Criteria → Quality gate: enforced by · Quality tests verify policy and required artifacts.
- Packaged criteria → Quality gate: enforced by · Packaged criteria are checked by the quality gate.
- Health check → Scenarios: validated by · Shared example behavior is covered by the scenario harness.
- Quality gate → CI: runs in · Quality checks are part of CI.
- Scenarios → Validate skill: runs in · Scenario checks are executed through the validation workflow.
- PR template → CI: requires evidence from · Pull requests must show validation proof.
- PR template → Validate skill: requires evidence from · Pull requests must show validation proof.
- Docs site → Deploy pages: published by · Static docs are deployed through GitHub Pages.
- Blog page → Docs site: part of · Blog page is one surface of the published docs site.
- Analytics → Docs site: supports · Analytics configuration supports the published docs surface.
- Agent config → SKILL.md: supports · Agent/tooling configuration complements skill authoring and usage.
- SKILL.md → Docs site: informs · Canonical guidance is reflected in published documentation.