orchestrator / tests
AGENTS.md1 starsChanged 6 months agoMITView on GitHub
Integration test suites for the orchestrator extension. 7 suites, 294 total assertions.
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- scripts/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- packagingSkill
The file
Files
tests/
Integration test suites for the orchestrator extension. 7 suites, 294 total assertions.
Running Tests
# All suites
for f in tests/test-s*.sh; do bash "$f"; done
# Individual suite
bash tests/test-s02-state-machine.shConventions
pass()/fail()functions with parallel indexed arrays (bash 3.2 compatible)- Structured
PASS:/FAIL:output with summary count - Exit 0 = all pass, exit 1 = any failure
- Fixtures in
tests/fixtures/— named by scenario
Suites
| File | Assertions | Covers |
| test-s01-structure.sh | 11 | Directory tree, config template |
| test-s02-state-machine.sh | 23 | 9 state derivations, config resolution, roadmap parsing, scaffolding |
| test-s03-design-artifacts.sh | 60 | 13 templates + 4 reference docs structural validation |
| test-s04-core-commands.sh | 51 | Verification scripts, dispatch scripts, 6 command files |
| test-s05-autonomous-mode.sh | 61 | Lifecycle scripts (lock, stuck, recovery, budget), auto/resume/discuss commands |
| test-s06-knowledge-lifecycle.sh | 57 | Knowledge scripts, rollback, consolidation, consolidate command |
| test-s07-integration.sh | 22 | Cross-slice manifest alignment, cross-references, capabilities, diagnostics |
Fixture Directory
Fixtures are minimal file trees that trigger specific states or scenarios. Key patterns:
state-*directories: one per state derivation result (9 total)verify-pass/verify-fail/verify-scope: verification scenariosdispatch-state: complete milestone tree for context assembly testingauto-*: lock, stuck, recovery, budget scenario fixturesknowledge-*,rollback-*,consolidate-*: knowledge/lifecycle fixtures
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.