Os Autoinst Distri Opensuse Skills — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Os Autoinst Distri Opensuse Skills (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
AI-powered developer productivity skills for the os-autoinst-distri-opensuse (OSADO) project. Provides modular Skills and Commands that act as an on-demand pair programmer aware of OSADO architecture and workflows.
[!IMPORTANT] The project moved from mpagot/os-autoinst-distri-opensuse-skills to os-autoinst/osado-skills. Installations from the old repository do not get updates any more. To migrate, remove the old installation and install again from the new repository, as described in Migration from the old repository.A few examples of what the assistant can do during a typical OSADO session.
You: I changed lib/sles4sap/azure_cli.pm -- what do I need to test?
AI: Your change touches 1 library file. Here's the verification plan:
- Unit test: prove -l -Ios-autoinst/ t/sles4sap/azure_cli.t
- Affected tests: tests/sles4sap/hana_install.pm, tests/sles4sap/azure_deploy.pm
- Schedules: schedule/sles4sap/hana_node.yaml
To get a clonable openQA job, confirm the target instance (osd or o3).You: I have the autoinst-log.txt from a failed job. The hana_install module
timed out -- was it a real failure or infrastructure lag?
AI: Running health check on autoinst-log.txt... 1 timeout detected in hana_install.
Running lag detection... backend loop count: 243,891 (threshold: 100,000).
Verdict: infrastructure stress on the worker host. The SUT likely had no issue.
Recommend: re-run the job on a different worker.You: I edited lib/sles4sap/azure_cli.pm and schedule/sles4sap/hana_node.yaml
-- what should I run locally?
AI: Analysing 2 changed files...
TIER 1 (instant):
perl -c lib/sles4sap/azure_cli.pm ... PASS
yamllint schedule/sles4sap/hana_node.yaml ... PASS
TIER 2 (before commit):
prove -lv -Ios-autoinst/ t/sles4sap/azure_cli.t ... 12 tests passed
make tidy ... no changes needed
All local checks passed. Ready to push.You: Add a documentation header to tests/sles4sap/qesap_ansible.pm
AI: Reading file... identified 3 openQA variables (ANSIBLE_PLAYBOOK, PUBLIC_CLOUD_PROVIDER,
QESAP_DEPLOYMENT_DIR) and 2 library dependencies (lib/sles4sap/qesapdeployment.pm,
lib/publiccloud/utils.pm).
Header generated from template. Running audit... PASS.
Metadata check... OK. POD whitespace... OK.You: I added a new function az_disk_create to lib/sles4sap/azure_cli.pm.
Generate a unit test for it.
AI: Scanning az_disk_create... 3 mandatory args (resource_group, name, size_gb),
1 optional (sku). Generating test skeleton with @calls capture pattern,
dies_ok for mandatory args, and separate subtest for the optional arg.
Written to t/21_sles4sap_azure_cli.t (appended 3 new subtests).
Run: prove -lv -Ios-autoinst/ t/21_sles4sap_azure_cli.tYou: /github_pr_create
AI: Inspecting staged changes... 2 files modified (lib/sles4sap/azure_cli.pm, t/sles4sap/azure_cli.t).
Proposed commit message:
"Add retry logic to azure_cli resource group creation"
Written to commit_message.txt -- please review and confirm.
You: Looks good, go ahead.
AI: Committed. Pushing to origin/add-retry-azure-cli...
Creating PR against os-autoinst/os-autoinst-distri-opensuse...
PR #18742 created: https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/18742For a complete list of use cases and detailed explanations, see Use Cases and Workflows.
| Skill | Description |
|---|---|
local-lint-test | Answers: "What's the fastest local command to validate my edit right now?" |
unit-test-wizard | Write and review unit tests for OSADO Perl library modules |
test-catalog | Add/audit Perldoc documentation headers on test modules |
vr-planner | Answers: "What remote openQA jobs should I clone to verify my change?" |
git-commit | Write OSADO-compliant commit messages for staged changes |
github-pr-create | End-to-end workflow: commit, push, and create PR upstream |
openqa-log-analyzer | Parse and extract sections from autoinst-log.txt |
openqa-log-fetcher | Fetch, cache, and inspect logs, test artifacts, and job metadata from openQA instances |
gh (GitHub CLI) for PR operations.jq for JSON processing.perl for compilation checks.uv / uvx (or openqa-log-local) for fetching openQA logs with openqa-log-fetcher.See the installation document.
# Clone and link for local development
git clone https://github.com/os-autoinst/osado-skills
cd osado-skills
gemini extensions link .For OpenCode skill testing within this repo, create a local symlink:
mkdir -p .opencode && ln -s ../skills .opencode/skillsChanges to skills, commands, and context files are picked up on the next Gemini CLI session.
# Run all local tests (installer + lint)
make test
# Run integration tests (requires podman or docker)
make test-integration
# Or with docker instead of podman
make test-integration CONTAINER_RT=docker
# Use a custom container image
IMAGE_REPO=my-registry.io/my-tester IMAGE_TAG=v2 make test-integrationIntegration tests run inside ghcr.io/mpagot/osado-gemini-tester:latest which ships with gemini, claude, and opencode pre-installed. The Makefile automatically pulls the latest image and verifies the digest against the remote registry before running.
To run the integration test script directly (bypassing the pull + digest check):
podman run --rm -v "$(pwd):/src:ro" \
ghcr.io/mpagot/osado-gemini-tester:latest /src/t/test_integration.shBoth IMAGE_REPO and IMAGE_TAG can be overridden via environment variables (the test script uses sensible defaults if unset).
Licensed under the same terms as OSADO (GPL-2.0-or-later).
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.