MCP server for querying AVEVA Historian data, providing tools to read tags, process values, analog summaries, and events
Four tools with complete input schemas and descriptions, but descriptions are minimal (10-60 chars, well below the 194-char baseline). All tools are READ_ONLY and follow verb_noun naming. However, output schemas are not documented in the visible code, parameter descriptions lack detail about formats/constraints, and no error handling guidance is provided. Tools are well-named and focused (single responsibility), but lack the depth needed for confident LLM selection and error recovery.
Retrieve tags from AVEVA Historian
Read analog summary statistics for specified tags from AVEVA Historian
Read events from AVEVA Historian
Read process values for specified tags from AVEVA Historian
Descriptions are too brief (10-60 chars vs 194-char baseline). 'Retrieve tags from AVEVA Historian' lacks context on when to use this vs other tools, what structure is returned, or prerequisites.
Output schemas are not documented in visible code. LLMs cannot plan downstream calls or extract required fields without knowing what get_tags returns (tag IDs, names, metadata structure, etc.).
Parameter descriptions lack format/constraint details. 'OData filter expression' is vague, no examples of valid syntax, no guidance on escaping, no error recovery hints if the filter is malformed.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 42 | <=2025-11-25 | v2 |
No error handling guidance. If a tag name is invalid or a time range is out of bounds, the LLM has no recovery path. Error responses should guide the agent to retry, adjust parameters, or call a discovery tool.
No pagination guidance in descriptions. read_process_values and read_analog_summary accept maxValues but do not document whether results are truncated, how to fetch the next batch, or what the default limit is.