Skills

A skill is a package of structured files that teaches an AI coding agent how to work with a specific tool or framework. The skill below was generated by Great Docs from this project’s documentation. Install it in your agent and it will be able to run commands, edit configuration, write content, and troubleshoot problems without step-by-step guidance from you.

Any agent — install with npx:

npx skills add https://github.com/NSLS2/recap/

Codex / OpenCode

Tell the agent:
Fetch the skill file at https://github.com/NSLS2/recap/skill.md and follow the instructions.

Manual — download the skill file:

curl -O https://github.com/NSLS2/recap/skill.md

Or browse the SKILL.md file.

SKILL.md

---
name: pyrecap
description: >
  Scientific framework for Reproducible Experiment Capture and Provenance. Use when writing Python code that uses the recap package.
license: BSD-3-Clause
compatibility: Requires Python >=3.10.
---

# Recap

Scientific framework for Reproducible Experiment Capture and Provenance

## Installation

```bash
pip install pyrecap
```

## API overview

### Client

Local and remote clients, namespace context, and permissions.

- `RecapClient`: Primary entry point for interacting with a RECAP provenance database
- `ActorPermissions`: Current actor's effective permissions returned by remote APIs
- `EffectivePermissions`
- `PermissionDecision`
- `DenialCode`

### Queries

Immutable query composition and field predicates.

- `QueryDSL`: Factory for namespace-scoped query families
- `Field`: Typed query-field expression builder
- `FieldPredicate`: Serializable field comparison used by :meth:`BaseQuery.where`
- `FieldOrdering`: Serializable ascending or descending field ordering
- `BaseQuery`: Immutable query base with filtering, pagination, loading, and execution
- `NamespaceQuery`: Query namespaces, including local and effective metadata filters
- `ResourceQuery`: Query resources, properties, templates, and parent relationships
- `ResourceTemplateQuery`: Query resource templates and optionally hydrate child relationships
- `ProcessTemplateQuery`: Query process templates and optionally hydrate steps and slots
- `ProcessRunQuery`: Query process runs with optional steps, resources, and parameters

### Builders

Context-managed resource and process builders.

- `ResourceTemplateBuilder`
- `ResourceBuilder`
- `ProcessTemplateBuilder`: Builder for process templates, resource slots, and step templates
- `StepTemplateBuilder`: Scoped editor for one step's parameter groups and resource bindings
- `ProcessRunBuilder`: Builder for process-run assignments, parameters, and child steps
- `AttributeGroupBuilder`

### Models and Lifecycle

Public schemas, lifecycle state, directions, and copy options.

- `LifecycleStatus`: Monotonic lifecycle state for persisted RECAP entities
- `Direction`: Direction of resource flow through a process step
- `validate_transition`: Validate lifecycle transition, allowing idempotent same-state updates
- `NamespaceRef`: Lightweight namespace identity used in relationships and query results
- `NamespaceContext`: Active namespace scope attached to client queries and writes
- `NamespaceSchema`: Persisted namespace with lifecycle, hierarchy, and metadata fields
- `ResourceTypeSchema`: A type tag associated with a resource or resource template
- `ResourceCopyChanges`: Property changes applied while copying a resource
- `ResourceCopyOptions`: Name, parent, and property overrides for a resource copy operation
- `ResourceTemplateRef`: Full blueprint for a category of resources
- `ResourceTemplateSchema`: Full blueprint for a category of resources
- `ResourceRef`: A concrete resource instance created from a :class:`ResourceTemplateSchema`
- `ResourceSchema`: A concrete resource instance created from a :class:`ResourceTemplateSchema`
- `ResourceAssignmentSchema`: Represents the assignment of a resource to a slot in a process run
- `ProcessTemplateRef`: Blueprint for a workflow, defining its ordered steps and resource slots
- `ProcessTemplateSchema`: Blueprint for a workflow, defining its ordered steps and resource slots
- `ProcessRunRef`: A concrete execution of a :class:`ProcessTemplateSchema`
- `ProcessRunSchema`: A concrete execution of a :class:`ProcessTemplateSchema`
- `StepTemplateRef`: Blueprint for a single workflow step within a process template
- `StepTemplateSchema`: Blueprint for a single workflow step within a process template
- `StepSchema`: A concrete step instance within a :class:`~recap.schemas.process.ProcessRunSchema`
- `ParameterSchema`: A parameter group instance attached to a step
- `AttributeTemplateSchema`: Persisted blueprint for a single typed attribute
- `AttributeGroupRef`: A named group of :class:`AttributeTemplateSchema` instances
- `AttributeGroupTemplateSchema`: A named group of :class:`AttributeTemplateSchema` instances
- `AttributeValueSchema`: A single stored attribute value with an optional physical unit
- `AttributeTemplateValidator`: Transient validator for a raw attribute definition supplied via the builder API

## Resources

- [Full documentation](https://github.com/NSLS2/recap/)
- [llms.txt](llms.txt) — Indexed API reference for LLMs
- [llms-full.txt](llms-full.txt) — Comprehensive documentation for LLMs
- [Source code](https://github.com/NSLS2/recap.git)