Delete a file. Default behaviour is a soft delete (the file goes to the trash and can be restored). Pass hard: true to permanently remove a file that's ALREADY in the trash — this also reclaims storage quota. [requires a signed-in account]
| Name | Required | Description | Default |
|---|---|---|---|
| hard | No | Hard-delete (only allowed on files already in trash). | |
| file_id | Yes |
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?
The description substantially expands on the annotations: it reveals the default soft-delete behavior, that files go to the trash and can be restored, that hard delete is only allowed on trashed files, and that it reclaims storage quota. This is valuable safety-relevant context beyond destructiveHint=true.
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?
The description is two tight sentences plus an auth note. It front-loads the core action, then explains the default, the conditional hard-delete path, and the side effect. Every sentence earns its place with no filler.
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 two-parameter destructive tool with no output schema, this description is complete: it names the required auth state, explains both parameter behaviors, gives the precondition for hard deletion, and discloses the storage-quota consequence. An agent has enough to invoke it correctly.
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?
The hard parameter is explained in the description beyond its schema text, including the 'already in trash' precondition and storage-quota effect. The file_id parameter has no schema description, but its meaning is obvious from the tool's purpose and parameter name. The description compensates reasonably for the 50% schema coverage.
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 starts with a specific verb and resource: 'Delete a file.' It clearly distinguishes the tool from sibling tools like octopus_read and octopus_move by focusing on deletion. The soft-delete vs. hard-delete distinction further sharpens the purpose.
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?
The description clearly explains the default behavior (soft delete) and when to use the hard delete option ('ALREADY in the trash'). It also notes that a signed-in account is required. It does not explicitly name alternative tools, but no sibling tool serves the same delete-file role, so the guidance is sufficient.
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.
Multiple tool pairs are near-identical: octopus_mkdir/octopus_make_folder and octopus_move/octopus_move_file are literal duplicates, analyze_hash/generate_hash both compute hashes, convert_word_to_pdf overlaps convert_document, and photo_compress/photo_compress_to_size plus pdf_thumbnails/pdf_to_images have fuzzy boundaries. The descriptions are detailed and cross-reference each other helpfully, but at 144 tools an agent will regularly misselect.
The dominant {category}_{verb}_{object} snake_case pattern (pdf_*, photo_*, convert_*, analyze_*, media_*) is largely consistent and predictable. However, outliers like chatwithyourpdf and describe_image break the category-prefix convention, and the octopus namespace mixes bare verbs (read, write, mkdir) with verb_noun forms (make_folder, move_file, search_meta) inconsistently.
144 tools is an extreme count for any MCP server. The broad scope (PDF, photo, video, audio, conversion, analysis, generation, file storage, web, e-sign) justifies some volume, but the count is inflated by batch and inspect variants (pdf_to_excel + batch + inspect), duplicate tools, and overlapping converters. An agent faces an overwhelming selection surface.
Per-domain coverage is remarkably deep: PDF spans merge/split/compress/protect/unlock/metadata/OCR/watermark and bidirectional conversion; photo covers editing, format conversion, face handling, OCR, and collage; file storage has full CRUD plus search. Minor gaps exist (no audio transcription, no video metadata editing, no deletion of PDF pages is actually covered via pdf_delete_pages) but the surface has no dead ends for its declared domains.
