AGENT SKILL

bridge_recovery_harness

Run a bounded BridgeStan score-model proxy recovery that invokes generated module scripts, logs command evidence, and writes validator-ready recovery artifacts.

License 许可
Apache-2.0
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
bridge_recovery_harness
description
Run a bounded BridgeStan score-model proxy recovery that invokes generated module scripts, logs command evidence, and writes validator-ready recovery artifacts.

02 资源文件 2

03 SKILL.md 正文

Bridge Recovery Harness

Use this skill when a soft-mode recovery must prove BridgeStan-like score-model behavior without compiling the full BridgeStan stack. It coordinates the other generated skills and produces auditable recovery artifacts.

Inputs

  • Attempt directory.
  • Generated skills root.
  • module_plan.json fast recovery target.
  • environment/runtime_handoff.json.

Outputs

  • recovery/recovery_result.json.
  • recovery/source_manifest.json.
  • recovery/logs/experiment_command_log.json.
  • recovery/logs/generated_skill_invocations.json.
  • Data, contract, transform, and score logs used by the proxy.

Workflow

  1. Confirm the runtime handoff recommends reduced/proxy recovery or that full BridgeStan runtime is unavailable.
  2. Create a tiny Bernoulli Stan source and observation data inside the current attempt.
  3. Call the generated contract, transform, and score scripts through subprocesses.
  4. Compute a mechanism-check pass-rate metric from the generated outputs.
  5. Write source manifest entries limited to allowed recovery sources.
  6. Run the Distiller recovery validator after the harness completes.

Validation

Run python tests/test_bridge_recovery_harness.py from this skill directory for a command-free smoke check of metric aggregation.

Limitations

This harness is a declared proxy. It must not claim that the real BridgeStan package or compiled Stan shared object ran unless environment evidence proves that separately.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-solve

END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-pattern-fusion

Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).

★ 351