gen-crash-test
Skill4 starsChanged 6 months agoApache-2.0View on GitHub
Crash-recovery test for the sagaflow generic interpreter. Multiple sequential tool calls with a sleep in the middle.
In this repository57 other files it gives its agents.
- autopilotSkill
- build-python-librarySkill
- build-temporalSkill
- canon-design-reviewSkill
- canon-pr-reviewSkill
- canon-reviewSkill
- ccr-modelsSkill
- ccr-runSkill
- chartSkill
- ci-health-reportSkill
- code-quality-trendsSkill
- competitive-matrixSkill
- councilSkill
- create-skillSkill
- deep-debug-ensemble-v1Skill
- deep-debugSkill
- deep-designSkill
- deep-ideaSkill
- deep-planSkill
- deep-qa-ensemble-v1Skill
- deep-qaSkill
- deep-researchSkill
- dependency-auditSkill
- deploy-status-reportSkill
- develop-board-gameSkill
- diagramSkill
- docxSkill
- dora-lite-reportSkill
- find-skillSkill
- flaky-test-diagnoserSkill
- gap-finderSkill
- gen-smokeSkill
- goal-managerSkill
- init-github-repoSkill
- loop-until-doneSkill
- magic-fetchSkill
- ml-pipeline-reportSkill
- monitorSkill
- napSkill
- pipeline-health-reportSkill
- platform-friction-detectorSkill
- pptxSkill
- prod-readinessSkill
- prompt-clash-ensembleSkill
- prompt-clashSkill
- retrospectSkill
- slack-digestSkill
- slidesSkill
- specSkill
- sprint-retroSkill
- tableSkill
- team-temporalSkill
- timelineSkill
- update-stop-gateSkill
- user-activity-reportSkill
- vc-gauntletSkill
- xlsxSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 27
Run sagaflow doctor. If Temporal is unreachable, stop immediately and report:
- Distinctive: line 31
Do NOT attempt an in-session workaround — this skill tests sagaflow, not Claude Code tools.
- Very specific: line 37
Use write_artifact with path step1.md content Step 1 complete for $name.
- Very specific: line 38
Use bash with command sleep 6 && echo 'slept 6s' and a 30s timeout.
- Distinctive: line 41
After step 3, end your turn.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
gen-crash-test
Crash-recovery test for the sagaflow generic interpreter's crash-recovery and sequential tool execution. It requires sagaflow — there is no in-session fallback because the test target is the sagaflow interpreter itself.
Pre-flight check
Run sagaflow doctor. If Temporal is unreachable, stop immediately and report:
gen-crash-test requires a running sagaflow worker (Temporal backend). Run
temporal server start-devandsagaflow worker run, then retry.
Do NOT attempt an in-session workaround — this skill tests sagaflow, not Claude Code tools.
Steps (sagaflow only)
Execute these three steps IN ORDER, one at a time, waiting for each tool result before the next:
- Use
write_artifactwith pathstep1.mdcontentStep 1 complete for $name. - Use
bashwith commandsleep 6 && echo 'slept 6s'and a 30s timeout. - Use
write_artifactwith pathstep3.mdcontentStep 3 complete for $name — all steps survived!.
After step 3, end your turn. One text summary, no more tool calls.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

