Senior Prompt Engineer
v2.1.1This skill should be used when the user asks to "optimize prompts", "design prompt templates", "evaluate LLM outputs", "build agentic systems", "implement RA...
⭐ 5· 1.7k·12 current·12 all-time
byAlireza Rezvani@alirezarezvani
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name, description, SKILL.md examples, and included scripts (prompt_optimizer.py, agent_orchestrator.py, rag_evaluator.py) are consistent with a 'senior prompt engineer' toolkit for analyzing prompts, RAG evaluation, and agent workflows. The skill does not declare unrelated env vars, binaries, or config paths. Minor note: the registry labeled this as 'instruction-only' while shipping multiple Python scripts — that is an internal inconsistency but not a functional mismatch.
Instruction Scope
SKILL.md instructs the agent/user to run the included Python scripts against local files (prompts, contexts, agent YAML). The visible scripts perform static analysis and local validation only. However rag_evaluator.py was truncated in the provided content; if it performs network calls (e.g., to embedding or LLM services) or looks up external endpoints, that behavior is not declared in SKILL.md or requires.env and should be inspected.
Install Mechanism
No install spec is present and the scripts are run directly with the Python interpreter. Nothing in the manifest attempts to download or extract remote archives or install third-party packages automatically.
Credentials
The skill declares no required environment variables or credentials and the visible code operates on local files. This is proportionate to the stated purpose. Caveat: if rag_evaluator.py (not fully shown) expects API keys or contains hardcoded endpoints, those would be out-of-band and should be verified before use.
Persistence & Privilege
The skill does not request always:true or any elevated/platform-wide privileges. It contains local utility scripts and reference docs only, and does not attempt to modify other skills or agent configuration beyond validating user-supplied agent config files (via agent_orchestrator).
Assessment
This skill appears to be a coherent prompt-engineering toolkit that runs local Python scripts to analyze prompts and agent configs. Before installing or running:
- Inspect rag_evaluator.py (the full file) for any network calls, hardcoded endpoints, or attempts to use API keys; the SKILL.md does not declare any required credentials.
- Run the scripts on non-sensitive sample files first and in a sandboxed environment (or virtualenv) so you can observe any unexpected outbound network traffic.
- Review any agent_config.yaml you give to agent_orchestrator.py—the tool validates and parses configs but uses a simple YAML parser; avoid feeding untrusted configs that could cause unexpected behavior.
If rag_evaluator.py uses external LLM/embedding services, ensure you supply credentials explicitly and understand where data is sent before using with private data.Like a lobster shell, security has layers — review code before you run it.
latestvk97atk8e2vzqbnjnfcktcb5kc982k9zf
License
MIT-0
Free to use, modify, and redistribute. No attribution required.