Geruyang/GAS-Agent-Skills / GAS-Agent-Skills/gas-centralized-development
gas-centralized-development
VerifiedUse when a user requests 集权模式, 集权开发模式技能, 指挥者模式, 中枢开发, or centralized development, or when coupled project work needs unified direction, subordinate supervision, and coordinated integration and release.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 5.0 KB |
| README.zh-CN.md | markdown | 7.7 KB |
| SKILL.md | skill md | 20.9 KB |
| agents/openai.yaml | config | 330 B |
| evals/runtime-scenarios.json | config | 4.9 KB |
| evals/scenarios.json | config | 41.8 KB |
| references/protocol.md | markdown | 45.3 KB |
| references/research-basis.md | markdown | 5.3 KB |
| references/runtime-evidence.md | markdown | 16.7 KB |
| scripts/gas_runtime.py | script | 22.9 KB |
| scripts/validate_skill.py | script | 23.5 KB |
| templates/command.example.json | config | 1.9 KB |
| templates/decision.example.json | config | 1.1 KB |
| templates/release.example.json | config | 2.0 KB |
| templates/review.example.json | config | 1.5 KB |
| templates/run.example.json | config | 7.1 KB |
| templates/runtime.example.json | config | 4.4 KB |
| templates/supervision.example.json | config | 3.0 KB |
| templates/task.example.json | config | 3.0 KB |
| tests/test_revision_contract.py | script | 3.6 KB |
| tests/test_role_update_contract.py | script | 7.6 KB |
| tests/test_skill.py | script | 6.1 KB |
| tests/test_v3_contract.py | script | 7.7 KB |
tests/test_v3_contract.py
View on GitHub"""Data-contract acceptance and negative cases for commander v3, not model behavior."""
import importlib.util
import json
import shutil
import tempfile
import unittest
from pathlib import Path
ROOT = Path(__file__).resolve().parents[1]
class CommanderContract(unittest.TestCase):
def setUp(self):
temporary = tempfile.TemporaryDirectory()
self.addCleanup(temporary.cleanup)
self.root = Path(temporary.name) / "skill"
shutil.copytree(ROOT, self.root, ignore=shutil.ignore_patterns("__pycache__"))
def doc(self, name):
return json.loads((self.root / name).read_text(encoding="utf-8"))
def edit(self, name, change):
path = self.root / name
obj = self.doc(name)
change(obj)
path.write_text(json.dumps(obj, ensure_ascii=False), encoding="utf-8")
def validate(self):
spec = importlib.util.spec_from_file_location("v3_checker", ROOT / "scripts/validate_skill.py")
checker = importlib.util.module_from_spec(spec)
spec.loader.exec_module(checker)
return checker.validate(self.root)
def rejected(self, fragment):
result = self.validate()
failures = [c["name"] for c in result["checks"] if not c["ok"]]
self.assertFalse(result["ok"])
self.assertTrue(any(fragment in c for c in failures), failures)
def test_three_subordinate_roles_and_commander_delivery(self):
run = self.doc("templates/run.example.json")
self.assertEqual(run["schema_version"], 3)
self.assertEqual(set(run["command_scope"]["roles"]), {"executor", "reviewer", "supervisor"})
self.assertEqual(run["delivery_responsibility"], {"owner_role": "coordinator", "separate_integrator_required": False})
def test_supervisor_covers_subordinates_and_human_covers_commander(self):
config = self.doc("templates/run.example.json")["supervision"]
self.assertEqual(set(config["target_roles"]), {"executor", "reviewer"})
self.assertEqual(config["reports_to"], "coordinator")
self.assertEqual(config["commander_supervised_by"], "human-only")
self.assertEqual(set(config["coverage_requirements"]), {"rule_compliance", "unsafe_actions", "instruction_alignment"})
def test_supervision_record_is_unrun_and_bound(self):
report = self.doc("templates/supervision.example.json")
self.assertEqual(report["schema_version"], 3)
self.assertEqual(report["record_type"], "supervision")
self.assertEqual(report["status"], "NOT_RUN")
self.assertEqual(report["findings"], [])
self.assertEqual(report["task_id"], self.doc("templates/task.example.json")["id"])
def test_missing_supervision_template_rejected(self):
path = self.root / "templates/supervision.example.json"
if path.exists():
path.unlink()
self.rejected("supervision:json")
def test_supervising_commander_rejected(self):
self.edit("templates/run.example.json", lambda d: d["supervision"]["target_roles"].append("coordinator"))
self.rejected("supervision:target-roles")
def test_supervisor_reporting_to_peer_rejected(self):
self.edit("templates/run.example.json", lambda d: d["supervision"].update(reports_to="reviewer"))
self.rejected("supervision:reports-to")
def test_nonhuman_commander_oversight_rejected(self):
self.edit("templates/run.example.json", lambda d: d["supervision"].update(commander_supervised_by="supervisor"))
self.rejected("supervision:human-oversight")
def test_ungranted_pause_authority_rejected(self):
self.edit("templates/run.example.json", lambda d: d["supervision"]["pause_authority"].update(granted=True))
self.rejected("supervision:pause-ungranted")
def test_supervisor_self_release_permission_rejected(self):
self.edit("templates/run.example.json", lambda d: d["supervision"].update(can_release=True))
self.rejected("supervision:limited-authority")
def test_reintroduced_integrator_rejected(self):
self.edit("templates/run.example.json", lambda d: d["delivery_responsibility"].update(owner_role="integrator"))
self.rejected("delivery:commander-owner")
def test_noncommander_delivery_recipient_rejected(self):
self.edit("templates/release.example.json", lambda d: d["command"].update(recipient_role="supervisor"))
self.rejected("delivery:commander-recipient")
def test_noncommander_delivery_execution_rejected(self):
self.edit("templates/release.example.json", lambda d: d["execution"].update(executor_role="integrator"))
self.rejected("delivery:commander-executor")
def test_fake_supervision_satisfaction_rejected(self):
self.edit("templates/decision.example.json", lambda d: d["acceptance_preconditions"].update(supervision_satisfied=True))
self.rejected("decision:supervision-default")
def test_invalid_supervision_config_rejected_without_crash(self):
self.edit("templates/run.example.json", lambda d: d.update(supervision=None))
self.rejected("supervision")
def test_empty_scenario_id_rejected(self):
self.edit("evals/scenarios.json", lambda d: d["cases"][0].update(id=""))
self.rejected("scenarios:valid-ids")
def test_null_scenario_rules_rejected_without_crash(self):
self.edit("evals/scenarios.json", lambda d: d["cases"][0].update(rules=None))
self.rejected("scenarios")
def test_malformed_markdown_is_reported_without_crash(self):
(self.root / "README.zh-CN.md").write_bytes(b"\xff\xfe\xff")
self.rejected("file:")
def test_nested_contract_objects_reject_scalars(self):
objects = {
"command": ("scope", "budget"), "release": ("target",),
"run": ("dispatch",), "task": ("budget", "test_plan", "dispatch", "integration", "release"),
}
for name, fields in objects.items():
path = self.root / f"templates/{name}.example.json"
original = path.read_bytes()
for field in fields:
for invalid in (None, [], True, "broken", 0):
with self.subTest(name=name, field=field, invalid=invalid):
path.write_bytes(original)
self.edit(path.relative_to(self.root), lambda d: d.update({field: invalid}))
self.rejected("object")
path.write_bytes(original)
def test_fabricated_completion_defaults_rejected(self):
fields = [
("release", ["preconditions", "independent_review_satisfied"]),
("release", ["preconditions", "platform_approvals_satisfied"]),
("review", ["required_checks_passed"]),
("task", ["verification", "independent_review_satisfied"]),
("decision", ["acceptance_preconditions", "platform_approvals_satisfied"]),
("run", ["runtime", "budget_enforcement_verified"]),
("run", ["runtime", "isolated_workspaces_verified"]),
]
for name, keys in fields:
path = self.root / f"templates/{name}.example.json"
original = path.read_bytes()
with self.subTest(name=name, keys=keys):
record = json.loads(original)
parent = record if len(keys) == 1 else record[keys[0]]
parent[keys[-1]] = True
path.write_text(json.dumps(record), encoding="utf-8")
self.assertFalse(self.validate()["ok"])
path.write_bytes(original)
def test_deploy_requires_target_verification_scope(self):
self.edit("templates/release.example.json", lambda d: d["preconditions"].update(verification_scope="accepted-source-tasks"))
self.rejected("delivery:verification-scope")
if __name__ == "__main__":
unittest.main()
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |