Skip to content

amazon-science/ammo / ai_cli_session/.codex/skills/ammo

Profile and optimize GPU kernels for vLLM inference on NVIDIA GPUs. Use when targeting a specific model, hardware, dtype, parallelism, and workload to improve latency.

FileKindSize
RUNTIME.mdmarkdown4.9 KB
SKILL.mdskill md18.5 KB
agents/ammo-auditor.mdmarkdown9.1 KB
agents/ammo-champion.mdmarkdown9.6 KB
agents/ammo-delegate.mdmarkdown3.2 KB
agents/ammo-implementer.mdmarkdown13.8 KB
agents/ammo-investigator.mdmarkdown4.0 KB
agents/ammo-report.mdmarkdown2.9 KB
agents/ammo-researcher.mdmarkdown10.3 KB
agents/ammo-resolver.mdmarkdown4.4 KB
agents/ammo-transcript-monitor.mdmarkdown7.8 KB
agents/openai.yamlexecutablescript313 B
data/ATTRIBUTION.mdmarkdown1.4 KB
data/gsm8k_full.jsonexecutablescript735.5 KB
data/gsm8k_subset.jsonexecutablescript116.8 KB
eval/EVAL-SKILL.mdexecutablescript21.8 KB
eval/agents/transcript_grader.mdexecutablescript18.0 KB
eval/causal/__init__.pyexecutablescript137 B
eval/causal/agents/causal_analyzer.mdexecutablescript10.5 KB
eval/causal/build_dag.pyexecutablescript21.6 KB
eval/causal/diff_versions.pyexecutablescript38.5 KB
eval/causal/extract_events.pyexecutablescript36.3 KB
eval/causal/generate_postmortem.pyexecutablescript17.0 KB
eval/causal/score_nodes.pyexecutablescript21.5 KB
eval/causal/templates/narrative_template.mdexecutablescript472 B
eval/causal/templates/viz_template.htmlexecutablescript9.5 KB
eval/causal/tests/__init__.pyexecutablescript151 B
eval/causal/tests/conftest.pyexecutablescript448 B
eval/causal/tests/fixtures/sample_session.jsonlexecutablescript10.4 KB
eval/causal/tests/fixtures/sample_session_data.jsonexecutablescript3.4 KB
eval/causal/tests/fixtures/sample_snapshot.jsonexecutablescript2.5 KB
eval/causal/tests/test_build_dag.pyexecutablescript2.7 KB
eval/causal/tests/test_diff_versions.pyexecutablescript13.5 KB
eval/causal/tests/test_extract_events.pyexecutablescript3.7 KB
eval/causal/tests/test_generate_postmortem.pyexecutablescript3.1 KB
eval/causal/tests/test_score_nodes.pyexecutablescript3.3 KB
eval/scripts/aggregate_versions.pyexecutablescript15.2 KB
eval/scripts/archive_run.pyexecutablescript9.1 KB
eval/scripts/codex_session.pyexecutablescript10.6 KB
eval/scripts/generate_dashboard.pyexecutablescript5.3 KB
eval/scripts/parse_artifacts.pyexecutablescript48.1 KB
eval/scripts/parse_session_logs.pyexecutablescript51.3 KB
eval/scripts/rescore_archived.pyexecutablescript5.5 KB
eval/scripts/score_campaign.pyexecutablescript43.8 KB
eval/scripts/snapshot_changes.pyexecutablescript11.1 KB
eval/scripts/verify_citations.pyexecutablescript8.8 KB
eval/tests/test_dashboard_bugs.pyexecutablescript8.8 KB
eval/tests/test_parse_session_logs.pyexecutablescript3.9 KB
eval/tests/test_parse_session_logs_round_inference.pyexecutablescript11.0 KB
eval/tests/test_score_campaign_diluted.pyexecutablescript9.4 KB
eval/viewer/dashboard_template.htmlexecutablescript15.1 KB
instrumentation/sitecustomize.pyexecutablescript3.6 KB
orchestration/audit-protocol.mdmarkdown9.3 KB
orchestration/debate-protocol.mdmarkdown11.2 KB
orchestration/integration-logic.mdmarkdown10.3 KB
orchestration/parallel-tracks.mdmarkdown8.9 KB
references/artifact-layout.mdexecutablescript22.4 KB
references/audit-invariants.mdmarkdown10.6 KB
references/champion-common-patterns.mdmarkdown3.4 KB
references/crossover-probing.mdexecutablescript4.7 KB
references/cudagraph-safety.mdmarkdown2.7 KB
references/debate-rules.mdmarkdown4.7 KB
references/debate-scoring-rubric.mdexecutablescript16.7 KB
references/e2e-delta-math.mdexecutablescript17.9 KB
references/e2e-latency-guide.mdexecutablescript9.6 KB
references/fusion-feasibility-heuristics.mdmarkdown5.6 KB
references/gpu-pool.mdexecutablescript5.4 KB
references/impl-track-rules.mdexecutablescript13.8 KB
references/nsys-profiling-guide.mdmarkdown9.8 KB
references/optimization-categories.mdexecutablescript11.0 KB
references/shutdown-protocol.mdexecutablescript3.1 KB
references/technology-selection.mdmarkdown5.0 KB
references/torch-compile-contract.mdmarkdown13.8 KB
references/validation-defaults.mdexecutablescript34.2 KB
references/workload-invocation.mdexecutablescript4.9 KB
references/writing-style.mdexecutablescript6.9 KB
scripts/ammo_state.pyexecutablescript142.0 KB
scripts/artifact_layout.jsonconfig16.3 KB
scripts/check_projection_accuracy.pyexecutablescript11.7 KB
scripts/create_evidence_template.pyexecutablescript9.7 KB
scripts/create_worktree_with_build.shexecutablescript10.2 KB
scripts/generate_validation_report.pyexecutablescript35.7 KB
scripts/gpu_reservation.pyexecutablescript27.1 KB
scripts/hook_cmd_classify.pyscript13.3 KB
scripts/mine_trace.pyscript29.4 KB
scripts/monitor_queue_ack.pyexecutablescript6.1 KB
scripts/ncu_sanity.shexecutablescript5.2 KB
scripts/ncu_sanity_driver.pyexecutablescript8.5 KB
scripts/new_target.pyexecutablescript22.0 KB
scripts/preflight_vllm_api.pyexecutablescript12.3 KB
scripts/reconcile_track_state.pyexecutablescript15.8 KB
scripts/remove_worktree_cleanup.shexecutablescript1.1 KB
scripts/resolve_codex_transcript.pyexecutablescript3.6 KB
scripts/run_vllm_bench_latency_sweep.pyexecutablescript249.6 KB
scripts/transcript_filter.pyexecutablescript38.4 KB
scripts/transitions.jsonexecutablescript1.5 KB
scripts/verify_validation_gates.pyexecutablescript88.8 KB
tests/agents/test-champion.mdexecutablescript20.4 KB
tests/agents/test-impl-champion.mdexecutablescript14.3 KB
tests/agents/test-implementer.mdexecutablescript15.8 KB
tests/agents/test-orchestrator.mdexecutablescript33.8 KB
tests/agents/test-researcher.mdexecutablescript21.0 KB
tests/agents/test-transcript-monitor.mdexecutablescript13.4 KB
tests/fixtures/diluted_pass/legacy_no_phase.golden.mdmarkdown1.1 KB
tests/fixtures/diluted_pass/legacy_no_phase.golden.summary.jsonconfig582 B
tests/fixtures/diluted_pass/legacy_no_phase.jsonconfig400 B
tests/fixtures/diluted_pass/legacy_scalar_phase.golden.mdmarkdown2.0 KB
tests/fixtures/diluted_pass/legacy_scalar_phase.golden.summary.jsonconfig1.0 KB
tests/fixtures/diluted_pass/legacy_scalar_phase.jsonconfig626 B
tests/fixtures/mining/real_campaign_bottleneck_analysis.mdmarkdown20.7 KB
tests/run_all.shexecutablescript1.6 KB
tests/scripts/test_transcript_filter.pyexecutablescript41.1 KB
tests/test_agent_runtime_contracts.pyexecutablescript1.8 KB
tests/test_ammo_state.pyexecutablescript71.7 KB
tests/test_artifact_layout_manifest.pyscript11.6 KB
tests/test_audit_started.pyscript18.2 KB
tests/test_correctness_comparator.pyscript12.6 KB
tests/test_diluted_pass.pyscript24.4 KB
tests/test_drift_compat.pyscript5.1 KB
tests/test_evidence_freshness.pyexecutablescript5.6 KB
tests/test_fail_closed.pyexecutablescript14.7 KB
tests/test_gpu_reservation.pyexecutablescript24.8 KB
tests/test_gpu_reservation_session_id.pyexecutablescript11.2 KB
tests/test_guidance_semantic_contract.pyscript24.9 KB
tests/test_hook_cmd_classify_parity.pyscript11.7 KB
tests/test_hook_semantics.pyexecutablescript95.3 KB
tests/test_mine_trace.pyscript24.2 KB
tests/test_mining_invalidated_schema.pyexecutablescript9.4 KB
tests/test_new_target_dp_ep.pyexecutablescript7.4 KB
tests/test_new_target_v2_scaffold.pyexecutablescript13.1 KB
tests/test_parse_artifacts.pyexecutablescript3.0 KB
tests/test_policy_convergence.pyscript797 B
tests/test_preflight_vllm_api.pyscript13.3 KB
tests/test_projection_accuracy_v2.pyexecutablescript4.1 KB
tests/test_reconcile_track_state.pyexecutablescript18.2 KB
tests/test_resolve_codex_transcript.pyexecutablescript3.1 KB
tests/test_state_schema_semantics.pyexecutablescript7.6 KB
tests/test_status_set_sync.pyexecutablescript9.1 KB
tests/test_sweep_baseline_from_arm.pyscript11.0 KB
tests/test_sweep_decode_capture.pyexecutablescript31.8 KB
tests/test_sweep_dp.pyexecutablescript45.4 KB
tests/test_sweep_fresh_cache.pyscript5.5 KB
tests/test_sweep_profiler_guard.pyscript22.6 KB
tests/test_sweep_real_prompts.pyscript20.1 KB
tests/test_sweep_refactor.pyexecutablescript37.6 KB
tests/test_sweep_round_slot.pyexecutablescript7.4 KB
tests/test_validation_report_phase_triage.pyexecutablescript10.0 KB
tests/test_validation_report_v2.pyexecutablescript6.2 KB
tests/test_verify_validation_gates.pyexecutablescript52.9 KB
tests/test_workload_matrix.pyexecutablescript19.3 KB

tests/test_preflight_vllm_api.py

View on GitHub
#!/usr/bin/env python3
# Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
# SPDX-License-Identifier: CC-BY-NC-4.0
"""Tests for preflight_vllm_api.py — the vLLM API coupling self-check.

The preflight must (1) run without a GPU and without building an engine,
(2) name every symbol the sweep script actually reads, and (3) FAIL when a
symbol is renamed. Point 3 is the whole value: a silent rename demotes the OTPS
denominator from decode-only to gross end-to-end.

These tests run without vLLM installed — the checks are driven through fake
modules injected into sys.modules, so the failure paths are exercised on any
machine.
"""

from __future__ import annotations

import dataclasses
import importlib.util
import sys
import types
from pathlib import Path

import pytest

_SCRIPTS_DIR = Path(__file__).resolve().parent.parent / "scripts"
_PREFLIGHT = _SCRIPTS_DIR / "preflight_vllm_api.py"
_SWEEP = _SCRIPTS_DIR / "run_vllm_bench_latency_sweep.py"


def _load_preflight():
    """Load a FRESH module instance (tests mutate its coupling tables)."""
    spec = importlib.util.spec_from_file_location("_preflight_under_test", _PREFLIGHT)
    mod = importlib.util.module_from_spec(spec)
    spec.loader.exec_module(mod)
    return mod


@pytest.fixture
def pf():
    return _load_preflight()


# ---------------------------------------------------------------------------
# Fake vLLM: every symbol the real coupling table names, at the right place.
# ---------------------------------------------------------------------------


@dataclasses.dataclass
class _FakeRequestStateStats:
    num_generation_tokens: int = 0
    scheduled_ts: float = 0.0
    first_token_ts: float = 0.0
    last_token_ts: float = 0.0


@dataclasses.dataclass
class _FakeEngineArgs:
    model: str = ""
    tensor_parallel_size: int = 1
    max_model_len: int = 0
    disable_log_stats: bool = False
    profiler_config: dict = dataclasses.field(default_factory=dict)
    max_num_batched_tokens: int = 0
    compilation_config: dict = dataclasses.field(default_factory=dict)
    attention_config: dict = dataclasses.field(default_factory=dict)
    structured_outputs_config: dict = dataclasses.field(default_factory=dict)
    data_parallel_size: int = 1
    pipeline_parallel_size: int = 1

    @classmethod
    def from_cli_args(cls, args):
        return cls()


@dataclasses.dataclass
class _FakeSamplingParams:
    n: int = 1
    temperature: float = 1.0
    top_p: float = 1.0
    ignore_eos: bool = False
    max_tokens: int = 16
    detokenize: bool = True
    stop: list = dataclasses.field(default_factory=list)
    seed: int = 0
    logprobs: int = 0


@dataclasses.dataclass
class _FakeBeamSearchParams:
    beam_width: int = 1
    max_tokens: int = 16
    ignore_eos: bool = False


@dataclasses.dataclass
class _FakeProfilerConfig:
    profiler: str = ""
    delay_iterations: int = 0
    max_iterations: int = 0
    torch_profiler_dir: str = ""


@dataclasses.dataclass
class _FakeVllmConfig:
    parallel_config: object = None
    profiler_config: object = None
    model_config: object = None


@dataclasses.dataclass
class _FakeParallelConfig:
    data_parallel_rank: int = 0


@dataclasses.dataclass
class _FakeModelConfig:
    max_model_len: int = 0


class _FakeLLM:
    def __init__(self, **kwargs):
        self.llm_engine = None

    def generate(self, *a, **k):
        return []

    def beam_search(self, *a, **k):
        return []

    def get_tokenizer(self):
        return None

    def get_metrics(self):
        return []

    def start_profile(self):
        pass

    def stop_profile(self):
        pass


class _FakeFlexibleArgumentParser:
    def __init__(self, *args, **kwargs):
        kwargs.pop("add_json_tip", None)
        self.kwargs = kwargs


_FAKE_LATENCY_SOURCE = '''
def add_cli_args(parser):
    parser.add_argument("--input-len")
    parser.add_argument("--output-len")
    parser.add_argument("--batch-size")
    parser.add_argument("--num-iters")
    parser.add_argument("--output-json")
    parser.add_argument("--num-iters-warmup")
    parser.add_argument("--use-beam-search")
    parser.add_argument("--disable-detokenize")
    parser.add_argument("--profile")


def save_to_pytorch_benchmark_format(args, results):
    pass
'''

_FAKE_GPU_WORKER_SOURCE = '''
def annotate():
    return "execute_context_0(0)_generation_8(8)"
'''

_FAKE_SPEC_METRICS_SOURCE = '''
NAMES = (
    "vllm:spec_decode_num_drafts",
    "vllm:spec_decode_num_draft_tokens",
    "vllm:spec_decode_num_accepted_tokens",
)
'''


def _module(name, **attrs):
    m = types.ModuleType(name)
    for k, v in attrs.items():
        setattr(m, k, v)
    return m


def _source_module(name, source, tmp_path):
    """Build a real importable module so inspect.getsource() works on it."""
    path = tmp_path / (name.replace(".", "_") + ".py")
    path.write_text(source, encoding="utf-8")
    spec = importlib.util.spec_from_file_location(name, path)
    mod = importlib.util.module_from_spec(spec)
    spec.loader.exec_module(mod)
    return mod


@pytest.fixture
def fake_vllm(monkeypatch, tmp_path):
    """Install a complete fake vLLM in sys.modules."""
    mods = {
        "vllm": _module(
            "vllm", LLM=_FakeLLM, SamplingParams=_FakeSamplingParams,
            __version__="9.9.9-fake",
        ),
        "vllm.benchmarks": _module("vllm.benchmarks"),
        "vllm.benchmarks.latency": _source_module(
            "vllm.benchmarks.latency", _FAKE_LATENCY_SOURCE, tmp_path
        ),
        "vllm.engine": _module("vllm.engine"),
        "vllm.engine.arg_utils": _module("vllm.engine.arg_utils", EngineArgs=_FakeEngineArgs),
        "vllm.inputs": _module("vllm.inputs", PromptType=dict),
        "vllm.sampling_params": _module(
            "vllm.sampling_params", BeamSearchParams=_FakeBeamSearchParams,
            SamplingParams=_FakeSamplingParams,
        ),
        "vllm.utils": _module("vllm.utils"),
        "vllm.utils.argparse_utils": _module(
            "vllm.utils.argparse_utils",
            FlexibleArgumentParser=_FakeFlexibleArgumentParser,
        ),
        "vllm.outputs": _module("vllm.outputs", RequestOutput=object),
        "vllm.v1": _module("vllm.v1"),
        "vllm.v1.metrics": _module("vllm.v1.metrics"),
        "vllm.v1.metrics.stats": _module(
            "vllm.v1.metrics.stats", RequestStateStats=_FakeRequestStateStats
        ),
        "vllm.distributed": _module("vllm.distributed"),
        "vllm.distributed.parallel_state": _module(
            "vllm.distributed.parallel_state", get_world_group=lambda: None
        ),
        "vllm.config": _module(
            "vllm.config", ProfilerConfig=_FakeProfilerConfig,
            VllmConfig=_FakeVllmConfig, ParallelConfig=_FakeParallelConfig,
            ModelConfig=_FakeModelConfig,
        ),
        "vllm.envs": _module(
            "vllm.envs",
            environment_variables={
                "VLLM_CACHE_ROOT": None,
                "VLLM_DISABLE_COMPILE_CACHE": None,
                "VLLM_EXECUTE_MODEL_TIMEOUT_SECONDS": None,
                "VLLM_WORKER_MULTIPROC_METHOD": None,
            },
        ),
        "vllm.v1.worker": _module("vllm.v1.worker"),
        "vllm.v1.worker.gpu_worker": _source_module(
            "vllm.v1.worker.gpu_worker", _FAKE_GPU_WORKER_SOURCE, tmp_path
        ),
        "vllm.v1.spec_decode": _module("vllm.v1.spec_decode"),
        "vllm.v1.spec_decode.metrics": _source_module(
            "vllm.v1.spec_decode.metrics", _FAKE_SPEC_METRICS_SOURCE, tmp_path
        ),
    }
    for name, mod in mods.items():
        monkeypatch.setitem(sys.modules, name, mod)
    return mods


# ---------------------------------------------------------------------------
# Green path
# ---------------------------------------------------------------------------


class TestPreflightPasses:
    def test_complete_api_surface_passes(self, pf, fake_vllm):
        assert pf.run_checks() == []

    def test_main_exits_zero_and_prints_version(self, pf, fake_vllm, capsys):
        assert pf.main([]) == 0
        out = capsys.readouterr().out
        assert out.startswith("OK:")
        assert "9.9.9-fake" in out

    def test_verbose_lists_each_check(self, pf, fake_vllm, capsys):
        pf.run_checks(verbose=True)
        out = capsys.readouterr().out
        assert "ok   RequestOutput.metrics.scheduled_ts" in out
        assert "ok   LLM.generate" in out


# ---------------------------------------------------------------------------
# Failure paths — one per check class
# ---------------------------------------------------------------------------


class TestPreflightDetectsRenames:
    def test_renamed_metrics_timestamp_fails(self, pf, fake_vllm):
        pf.REQUEST_METRICS_FIELDS = ["scheduled_ts", "first_token_time"]
        fails = pf.run_checks()
        assert any("first_token_time" in f for f in fails)

    def test_missing_top_level_symbol_fails(self, pf, fake_vllm):
        pf.MODULE_ATTRS = dict(pf.MODULE_ATTRS)
        pf.MODULE_ATTRS["vllm"] = ["LLM", "GoneSymbol"]
        assert any("vllm.GoneSymbol" in f for f in pf.run_checks())

    def test_missing_module_fails(self, pf, fake_vllm):
        pf.MODULE_ATTRS = dict(pf.MODULE_ATTRS)
        pf.MODULE_ATTRS["vllm.nope.gone"] = ["Thing"]
        assert any("vllm.nope.gone" in f for f in pf.run_checks())

    def test_renamed_engine_args_field_fails(self, pf, fake_vllm):
        pf.ENGINE_ARGS_FIELDS = list(pf.ENGINE_ARGS_FIELDS) + ["renamed_field"]
        assert any("EngineArgs.renamed_field" in f for f in pf.run_checks())

    def test_removed_latency_flag_fails(self, pf, fake_vllm):
        pf.LATENCY_CLI_FLAGS = ["--output-len", "--retired-flag"]
        assert any("--retired-flag" in f for f in pf.run_checks())

    def test_removed_env_var_fails(self, pf, fake_vllm):
        pf.ENV_VARS = ["VLLM_CACHE_ROOT", "VLLM_RETIRED"]
        assert any("VLLM_RETIRED" in f for f in pf.run_checks())

    def test_changed_nvtx_marker_fails(self, pf, fake_vllm):
        pf.NVTX_MARKER_TOKENS = ["_generation_", "_new_marker_"]
        assert any("_new_marker_" in f for f in pf.run_checks())

    def test_renamed_spec_counter_fails(self, pf, fake_vllm):
        pf.SPEC_DECODE_COUNTERS = ["vllm:spec_decode_num_drafts", "vllm:renamed"]
        assert any("vllm:renamed" in f for f in pf.run_checks())

    def test_rejected_parser_kwarg_fails(self, pf, fake_vllm, monkeypatch):
        class _StrictParser:
            def __init__(self, *a, **k):
                raise TypeError("unexpected keyword argument 'add_json_tip'")

        monkeypatch.setitem(
            sys.modules, "vllm.utils.argparse_utils",
            _module("vllm.utils.argparse_utils", FlexibleArgumentParser=_StrictParser),
        )
        assert any("FlexibleArgumentParser" in f for f in pf.run_checks())

    def test_main_exits_one_and_lists_failures(self, pf, fake_vllm, capsys):
        pf.REQUEST_METRICS_FIELDS = ["gone_ts"]
        assert pf.main([]) == 1
        out = capsys.readouterr().out
        assert out.startswith("FAIL:")
        assert "gone_ts" in out

    def test_main_exits_two_without_vllm(self, pf, monkeypatch, capsys):
        monkeypatch.setitem(sys.modules, "vllm", None)
        assert pf.main([]) == 2
        assert "cannot import vllm" in capsys.readouterr().out


# ---------------------------------------------------------------------------
# The table must stay in sync with the sweep script, and stay GPU-free.
# ---------------------------------------------------------------------------


class TestCouplingTableMatchesSweep:
    @pytest.fixture(scope="class")
    def sweep_source(self):
        return _SWEEP.read_text(encoding="utf-8")

    def test_metrics_timestamps_match_the_harvest(self, pf, sweep_source):
        """_request_phase_deltas is the single owner; preflight must name its fields."""
        harvest = sweep_source[sweep_source.index("def _request_phase_deltas"):]
        harvest = harvest[:harvest.index("def _p50")]
        for field in pf.REQUEST_METRICS_FIELDS:
            assert field in harvest, f"{field} is not read by _request_phase_deltas"

    def test_every_sweep_vllm_import_is_covered(self, pf, sweep_source):
        import re

        for mod, names in re.findall(
            r"^\s+from (vllm[\w.]*) import ([\w, ]+)", sweep_source, re.M
        ):
            covered = pf.MODULE_ATTRS.get(mod, [])
            for name in [n.strip() for n in names.split(",")]:
                base = name.split(" as ")[0].strip()
                # Two import forms: `from vllm import LLM` (attribute of mod)
                # and `from vllm.benchmarks import latency` (submodule).
                assert base in covered or f"{mod}.{base}" in pf.MODULE_ATTRS, (
                    f"sweep imports {mod}.{base}; preflight does not check it"
                )

    def test_env_vars_match_the_sweep_injectors(self, pf, sweep_source):
        for name in pf.ENV_VARS:
            assert name in sweep_source, f"{name} is no longer injected by the sweep"

    def test_spec_counters_match_the_snapshot(self, pf, sweep_source):
        for name in pf.SPEC_DECODE_COUNTERS:
            assert name in sweep_source

    def test_preflight_never_builds_an_engine(self, pf):
        """No-GPU contract: the script must not instantiate LLM or call generate."""
        src = _PREFLIGHT.read_text(encoding="utf-8")
        assert "LLM(" not in src
        assert ".generate(" not in src
        assert "import torch" not in src
        assert "add_cli_args(" not in src  # builds pydantic configs; needs a device
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: 198kUpdated Claude CodeCodexVerifiedInstall
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: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
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: 194kUpdated Claude CodeCodexVerifiedInstall