Menu
Run when the user asks to investigate, troubleshoot, or fix a flaky Go test.
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/gravitational/teleport --skill flaky-testThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Showing SKILL.md
| name | flaky-test |
| description | Run when the user asks to investigate, troubleshoot, or fix a flaky Go test. |
You are debugging a flaky test. In Teleport, flaky tests often pass locally on powerful developer machines, but then fail on constrained CI/CD hardware.
Sometimes the problem is due to the way the test is written, but there are plenty of occaisions where the flaky test is highlighting a problem in the implementation itself.
go test -c -o flaky.test .stress tool:
stress -failfast -p 30 ./flaky.test -test.run TestNameHere.
If stress is not installed it can be installed with
go install golang.org/x/tools/cmd/stress@latest.
You may need to increase the -p (parallelism) flag in order to reproduce
the flake. Avoid increasing the parallelism too much. Anything above -p 100
starts to induce failures in common setup logic rather than in the test itself.stress until it passes consistently.flaky.test binary. Do not make any commits,
the developer will want to review your changes and make adjustments.Prefer deterministic changes like channel-based notifications or
testing/synctest over retry-based solutions like require.Eventually.
Extending the timeout of a test is usually an anti-pattern. Try to fix the bug
in the implementation or rewrite the test to run faster.
Another way to make tests more reliable is to make them do less. Look for opportunities to improve the runtime of the test. This might mean optimizing the underlying code, better isolating the feature under test, or disabling features that are enabled by default but otherwise unnecessary for the feature being tested.
Use when reviewing who can access which resources in a Teleport cluster, or whether that access is actually used — access list / ACL recertification, "who can reach this database/server", "what can this user access", attesting access for audit/compliance, and finding dormant or unused standing privileges. Covers `tctl access-review`, the `access_path` SQL query language, output semantics, and combining it with `tctl investigate`.
Use when investigating identity or resource activity in a Teleport cluster. Covers `tctl investigate` filter semantics, fast facet-only exploration, custom Lucene query construction, and token-efficient parsing patterns.
Use for Teleport access list work with tctl; listing available resources (servers, databases, apps, etc.) before choosing an access grant target, creating new access lists, updating existing lists, or deleting/retiring lists. Trigger for requests like "give alice access", "create an access list", "standing access", "Access Request", "custom access list", "show/list apps", "show AWS IC permission sets", "add AWS IC to this list", "change owners/members/access", "remove access", "delete/retire/tear down an access list". Handles access-type lists where Teleport creates supporting roles, and custom lists that grant existing roles/traits.
Based on SOC occupation classification
