analyze-text
Use Presidio Analyzer for text PII/PHI detection, custom recognizers, NLP engines, score thresholds, allow lists, and YAML-configured integrations.
- License 许可
- MIT
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- analyze-text
- license
- MIT
- metadata
- { "disco-role": "operating" }
- description
- Use Presidio Analyzer for text PII/PHI detection, custom recognizers, NLP engines, score thresholds, allow lists, and YAML-configured integrations.
- disable-model-invocation
- true
02 资源文件 6
- 参考文档references/api-reference.md5.1 KB
- 参考文档references/recognizers-and-nlp.md5.5 KB
- 参考文档references/supported-entities-and-languages.md3.7 KB
- 参考文档references/troubleshooting.md4.5 KB
- 脚本scripts/analyzer_smoke.py3.6 KB
- 脚本scripts/custom_recognizer_smoke.py3.7 KB
03 SKILL.md 正文
analyze-text
Use this sub-skill for free-text Presidio Analyzer work:
- single-text or batch PII/PHI detection
- custom pattern, deny-list, and remote recognizers
- supported entities, languages, country filters, and context words
- score thresholds, allow lists, and ad hoc recognizers
- NLP engine selection and model configuration
- optional GLiNER, LangExtract, Azure AI Language, and AHDS routing
Route elsewhere
- Text anonymization and deanonymization → sibling anonymize-text skill
- DataFrame / JSON structured workflows → sibling structured-data skill
- Image OCR / bbox redaction → sibling image-redaction skill
- CLI file or directory scans → sibling cli-scans skill
- HTTP service startup or deployment → root service reference, not this sub-skill
Primary surface
AnalyzerEngineBatchAnalyzerEngineRecognizerRegistryPatternPatternRecognizerRecognizerResultAnalyzerEngineProviderNlpEngineProviderNoOpNlpEngine
Preferred workflow
- Pick
AnalyzerEnginefor one text orBatchAnalyzerEnginefor many values. - Load recognizers into a
RecognizerRegistry. - Choose an NLP engine:
- default spaCy path
NoOpNlpEnginefor no-download, pattern-only flows- Stanza / Transformers / GLiNER / LangExtract / Azure AI Language / AHDS when configured
- Call
analyze()with entities, thresholds, context, allow lists, or ad hoc recognizers.
Safe bundled checks
scripts/analyzer_smoke.pyscripts/custom_recognizer_smoke.py
Reference notes
references/api-reference.mdreferences/recognizers-and-nlp.mdreferences/supported-entities-and-languages.mdreferences/troubleshooting.md
Fast troubleshooting
- Missing default spaCy model: install the documented default model or use
NoOpNlpEnginewith custom recognizers. - Unsupported language or unknown entity: confirm the registry, NLP engine, and entity name all match.
- Threshold or allow-list surprise: check request-level threshold precedence and exact vs regex allow-list matching.
- Regex timeout or flags: review
global_regex_flags, requestregex_flags, andREGEX_TIMEOUT_SECONDS. - Optional extra missing: install only the extra needed for the configured recognizer or NLP engine.
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).