execution-caching
Provide execution caching workflows to reduce repeated runs and costs. Use when discussing cached execution, cache invalidation, or cache system operations.
Best use case
execution-caching is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Provide execution caching workflows to reduce repeated runs and costs. Use when discussing cached execution, cache invalidation, or cache system operations.
Teams using execution-caching should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/execution-caching/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How execution-caching Compares
| Feature / Agent | execution-caching | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Provide execution caching workflows to reduce repeated runs and costs. Use when discussing cached execution, cache invalidation, or cache system operations.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Execution Caching
## Rule Anchor
- `AGENTS.md` > "No Fallback Policy"
- `AGENTS.md` > "Development Patterns"
## Scope
Use this skill to design or operate execution caching systems for repeatable runs.
## Cache Structure (Reference)
```typescript
interface CacheEntry {
timestamp: number;
sourceFileHash: string;
dependencyVersions: Record<string, string>;
executionResult: {
nodes: WorkflowNode[];
edges: WorkflowEdge[];
logs: string[];
metadata: ExecutionMetadata;
};
success: boolean;
errorMessage?: string;
}
```
## Core Workflow
1. Compute a cache key from input and dependencies.
2. Load cache and validate integrity.
3. If valid and fresh, reuse cached result.
4. Otherwise run execution and save results.
## Cache Invalidation
- Source file content changes
- Dependency version changes
- Cache age exceeds policy
## Cache Operations (Examples)
- Clear cache for a specific execution key
- Clear all cache entries
- Show cache stats
- Force cached run when inputs are unchanged
## Log Analysis (Optional)
- Save full execution logs to files.
- Filter logs for specific patterns without re-running.
## No-Fallback Handling
- If cache integrity validation fails, stop execution and surface an error.
- Do not auto-run LLM execution as a fallback.
- Allow manual re-run with an explicit LLM mode by the user.