ecc-clickhouse-io
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a vulnerability surface for indirect prompt injection and SQL injection in its documentation of data ingestion patterns.
- Ingestion points: External data sources processed by the
bulkInsertTradesandstreamInsertsfunctions inreferences/upstream-summary.md. - Boundary markers: None provided; the workflow does not instruct the agent to use delimiters or ignore instructions embedded within the ingested data.
- Capability inventory: The skill includes database query execution and data movement capabilities using the
clickhouselibrary. - Sanitization: The TypeScript examples demonstrate manual string interpolation for SQL values (e.g.,
'${trade.id}'), which is an unsafe practice that does not perform proper escaping or use prepared statements. - [COMMAND_EXECUTION]: The workflow in
SKILL.mdinstructions the agent to execute fallback scripts from a local path (openclaw-ecc/scripts/) when primary tool hooks are unavailable. This execution pattern relies on the integrity of the files present in that specific local directory.
Audit Metadata