Closed intelfisz closed 1 year ago
Which version of scorecard are you using? With the latest scorecard, https://github.com/opiproject/opi-intel-bridge scores 10/10 on SAST. You're right that CodeQL is not detected (we don't parse GitHub composite actions or transitive action steps) but we should detect the existence of CodeQL as a check run on PRs.
Starting [SAST]
Finished [SAST]
RESULTS
-------
Aggregate score: 10.0 / 10
Check scores:
|---------|------|--------------------------------|--------------------------------|-----------------------------------------------------------------|
| SCORE | NAME | REASON | DETAILS | DOCUMENTATION/REMEDIATION |
|---------|------|--------------------------------|--------------------------------|-----------------------------------------------------------------|
| 10 / 10 | SAST | SAST tool is run on all | Info: all commits (30) are | https://github.com/ossf/scorecard/blob/main/docs/checks.md#sast |
| | | commits | checked with a SAST tool Warn: | |
| | | | CodeQL tool not detected | |
|---------|------|--------------------------------|--------------------------------|-----------------------------------------------------------------|
Please see https://github.com/ossf/scorecard/pull/3591, GitHub changed a string we were using for CodeQL detection. Make sure you're using Scorecard v4.13.1 or Scorecard Action v2.3.1
Thanks for the clarifications - indeed, this seems to be related to the recent change and works fine with the latest version. I am closing the issue.
Describe the bug In opi-intel-bridge repo I have run CodeQL scans for a long time - see https://github.com/opiproject/opi-intel-bridge/blob/main/.github/workflows/codeql.yml. The action reuses action defined in actions repo - see https://github.com/opiproject/actions/blob/main/.github/workflows/codeql.yml. However, OSSF score shows 0/10 points since it can't detect that CodeQL is actually run.
Reproduction steps Steps to reproduce the behavior:
Expected behavior I expect that the evaluation has more intelligence to detect factual CodeQL scan runs since it is a common practise to reduce code duplication in an organization and reuse actions between different repos.
Additional context It seems that a change has happened recently or SAST was introduced as this score penalty was not observed before and we run CodeQL scans for a long time already.