c-api
Guides agents locating Qiskit C headers and library, using qiskit.capi ctypes bindings, and handling public C-API build/link issues.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- c-api
- license
- Apache 2.0
- metadata
- { "disco-role": "operating" }
- description
- Guides agents locating Qiskit C headers and library, using qiskit.capi ctypes bindings, and handling public C-API build/link issues.
- disable-model-invocation
- true
02 资源文件 2
- 参考文档references/build-and-link.md1.5 KB
- 参考文档references/troubleshooting.md1.7 KB
03 SKILL.md 正文
Qiskit C API workflows
Use this sub-skill when the task involves Qiskit's public C API from Python: locating headers, locating the shared library, using ctypes bindings, planning downstream extension builds, or diagnosing source-build/link failures.
Read next
references/build-and-link.mdforqiskit.capi.get_include(),get_lib(), ctypes access, and source-build notes.references/troubleshooting.mdfor missing headers/library, Rust extension, stale editable install, and direct-link caveats.../../references/installation.mdfor source-build prerequisites.../../scripts/check_qiskit_environment.py --sections capifor a source-free C-API path check.
Include here
qiskit.capi.get_include()andqiskit.capi.get_lib().- Python-space ctypes bindings exposed under
qiskit.capi. - Downstream C-extension or standalone C-library build planning that uses Qiskit's public headers.
- Source-build environment variables and Rust-backed extension diagnostics.
Exclude or route elsewhere
- Python circuit construction belongs in
../circuit/SKILL.md. - Python transpilation APIs belong in
../transpiler/SKILL.md. - Backend/provider abstractions belong in
../providers/SKILL.md. - Generic C or Rust build-system advice not involving Qiskit belongs to a broader build-system skill.
Default route
Start here when the user mentions qiskit.capi, get_include, get_lib, qiskit.h, libqiskit, ctypes wrappers, C extension modules, make c, Rust build failures, or missing _accelerate during a Qiskit source install.
What to remember
- The Python package contains the public C headers and the shared-object library that backs the C API.
get_lib()is useful for local introspection, but direct linking is not the safest path for distributable Python extensions unless the user understands the ABI caveats.- Source builds require the supported Rust toolchain and the package build backend; normal wheels should already contain the built extension.
04 同仓库更多 Skills
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.
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.
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.
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.
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.
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).