The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script (md2html.py) using python3 at a specific workspace path (/root/.openclaw/workspace/deer-flow/skills/public/github-deep-research/scripts/md2html.py) for report formatting.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection, where malicious instructions embedded in external web content could influence the agent's behavior.
Ingestion points: Data is collected from untrusted external websites via web_search and web_fetch, and from internal sources via feishu_search_doc_wiki and feishu_bitable_app_table_record.
Boundary markers: Absent. The instructions do not define delimiters or "ignore embedded instructions" warnings for the processed data.
Capability inventory: The agent can perform web operations, access internal documents, write local files, and execute shell commands.
Sanitization: Absent. There are no requirements to sanitize, validate, or escape the content retrieved from the web before it is analyzed or included in reports.

company-research