AGENT SKILL

workspace-boundaries

Manage Potpie pots, repo defaults, and source registration boundaries.

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

01 前言 Front Matter

name
workspace-boundaries
license
Apache 2.0
metadata
{ "disco-role": "operating" }
description
Manage Potpie pots, repo defaults, and source registration boundaries.
disable-model-invocation
true

02 资源文件 2

03 SKILL.md 正文

Potpie workspace boundaries

Use this sub-skill when the task is about which Potpie workspace (pot) owns a repository, which source is registered, or why a command cannot find the expected repo default.

Read this when

  • The user asks about potpie pot ..., potpie source ..., default pot bindings, or source status.
  • A workflow needs to bind a local repo, remote repo identity, or provider source before graph operations.
  • A failure says there is no active pot, no linked repo, an unknown source, or ambiguous repository identity.

Do not use this for

  • Runtime/daemon setup: read ../runtime/SKILL.md.
  • Provider login, tokens, OAuth, or ledger binding: read ../auth-integrations/SKILL.md.
  • Graph read/write semantics: read ../graph-read/SKILL.md or ../graph-write/SKILL.md.
  • Bundled agent-skill installation: read ../skills-management/SKILL.md.

Operating procedure

  1. Inspect the active workspace with potpie pot list, potpie pot info, and potpie pot linked before creating a new pot.
  2. Register a repository source with potpie source add repo ... or potpie pot create ... --repo ... when the user wants Potpie to know about a codebase.
  3. Remember that source registration is not ingestion. It records source metadata so later graph, ledger, and nudge paths can reason about it.
  4. Use potpie pot default show|set|clear to manage the repo default instead of relying on the current shell directory.
  5. Avoid destructive commands such as potpie pot reset unless the user explicitly wants to clear data for that pot.

References

  • references/workflow.md — pot/source command matrix, repo-default resolution, and safe examples.
  • references/troubleshooting.md — repo identity, source normalization, no-scan behavior, and reset cautions.

Verification notes

  • Safe native candidates include pot creation, source CLI contract, repo-location, first-pot setup, empty-pot guidance, and source-resolution tests.
  • No accelerator backend is required for workspace boundary behavior.

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