Strip and replace PII/privileged entities per GDPR, EU AI Act, HIPAA, or a custom scheme. Requires tools.text (text) or tools.file (file). Returns anonymized document + entity map, or a generic permission denial.
| Name | Required | Description | Default |
|---|---|---|---|
| text | No | Inline text when no file_path | |
| regime | Yes | ||
| file_path | No | Absolute path to pdf/docx/csv/xlsx | |
| custom_entities | No | Entity types when regime is CUSTOM |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the burden of behavioral disclosure. It does disclose the return output ('anonymized document + entity map') and a fallback ('generic permission denial'), but it omits whether the original file is mutated, what permissions are needed, and any data-handling implications for PII.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three sentences front-load the core action, then the input requirement, then the output. There is no filler or repetition.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with no output schema and no annotations, the description covers the essential input/output flow but leaves out sibling differentiation, side-effect information, and error handling beyond a permission denial. Adequate but not complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 75%, and the description adds the meaningful relationship that text and file_path are alternatives ('Requires tools.text (text) or tools.file (file)'), which is not expressed as a schema constraint. This increment over the schema justifies above-baseline.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific action ('Strip and replace') and a specific resource ('PII/privileged entities'), and enumerates the regulatory schemes supported. It does not explicitly contrast with the sibling 'redact_pii', so it misses the top score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is given for choosing this tool over the sibling 'redact_pii' or 'list_allowed_commands'. The only usage-related sentence states the input requirement ('Requires tools.text or tools.file'), which is a constraint rather than a selection criterion.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
anonymize_document and redact_pii heavily overlap: both remove/replace PII from text, with no clear boundary or explicit contrast between them. list_allowed_commands is unrelated to the core privacy workflow, so agents may struggle to choose correctly.
All tool names follow a clean snake_case verb_noun pattern: anonymize_document, list_allowed_commands, redact_pii. The naming is predictable and consistent, even though two tools overlap in purpose.
Three tools is a reasonable, compact surface for a narrow privacy/redaction server. The count is slightly thin for a broad privacy domain like GDPR/HIPAA, but not unreasonable.
Core anonymization and redaction are covered, but there is no entity discovery, scheme management, or support for updating/removing previously created anonymization mappings. redact_pii also only handles text input, leaving file-based redaction entirely dependent on the overlapping anonymize_document tool.
