48h-Expert-Methodology
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only learning skill is coherent and purpose-aligned, but users should notice that it may perform research, enforce JSON-only output, and save generated results for later skill use.
This skill appears safe to install as an instruction-only research and study-formatting aid. Before using it, be aware that it is designed to produce rigid JSON, may perform source retrieval, and will save generated results to a local OpenClaw temporary path for later use.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may look up external sources to build the study output.
The skill may cause the agent to use search or retrieval tools. This is expected for a research methodology skill and is bounded to authority-source gathering.
Execute targeted retrieval of "Foundational Textbooks," "Peer-Reviewed Research," and "Academic Syllabi."
Use it for research tasks where external source retrieval is acceptable, and review the cited sources for relevance and reliability.
Responses may be rigid JSON instead of natural-language explanations.
The skill imposes strict output-format instructions. They are purpose-aligned with schema interoperability, but they can override a user's expectation of a normal explanatory answer while the skill is invoked.
System SHALL NOT output unstructured prose... No conversational filler before or after the JSON payload.
Invoke this skill when structured output is desired; ask separately for a human-readable explanation if needed.
Incorrect, sensitive, or low-quality generated study notes could persist and influence later audit workflows.
The skill persists generated research output for reuse by later skills. This is disclosed and purpose-aligned, but saved generated context can be reused or over-trusted later.
Write the final JSON to `~/.openclaw/swarm_tmp/expert_output.json`... Every output node MUST be referenceable by subsequent audit skills.
Review the saved JSON before relying on it in other skills, and delete or replace the file when it is no longer needed.