Server Quality Checklist

Tool Scores

• envi_classic_launchB
  3.3/5.0
  Behavior3/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  The description discloses key side effects: it creates a new evidence directory and launches a session, which the annotations do not convey. It also clarifies that only the session ID is returned. However, it omits details such as failure behavior, timeout semantics, or what 'controlled' means.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is a single, front-loaded sentence. It conveys the core purpose and return value with no filler or redundant details.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a tool with 4 parameters, side effects, and no schema property descriptions, this description is too sparse. It lacks parameter semantics, usage flow, prerequisites, and failure behavior. The return value is stated, but that does not compensate for the other missing context.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, so the description must clarify parameters. It only alludes to 'specified experiment workspace', which loosely maps to workspace_root and experiment_name, but provides no meaning for mode or timeout_seconds. This is minimal compensation for the lack of schema descriptions.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the action: create a fresh evidence directory and launch ENVI Classic 5.6. It also specifies the return value (session ID), which distinguishes this launch tool from sibling tools like invoke_menu, health_check, or close_session.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No explicit guidance on when to use this tool versus alternatives. It does not mention that it should be the initial step before other ENVI session tools, nor does it describe prerequisites or exclusions. The usage context is only implied by the sibling tool list.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_list_windowsA
  3.7/5.0
  Behavior3/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  Annotations already declare readOnlyHint, idempotentHint, and non-destructive behavior. The description adds that only 'visible top-level' windows are listed and specifies the data fields (titles, geometry), which is useful but not extensive.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  A single, focused sentence that front-loads the purpose and includes no extraneous information. Every word earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a simple read-only list operation with only one parameter and an output schema present, the description covers the core functionality. It adds the 'visible top-level' detail, which is helpful, and no complex prerequisites or errors are likely.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  The schema has no description for session_id, and the description only refers to 'specified controlled ENVI process' without explaining the parameter format, type, or expected values. With 0% schema coverage, this is insufficient.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool lists visible top-level windows, titles, and geometry for a specified ENVI process. The verb 'list' and specific resource make it distinct from sibling tools like launch, close, or invoke_menu.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Usage is implied: use when you need to inspect the windows of an ENVI session. No explicit alternatives or exclusions are provided, but the context makes the purpose fairly obvious.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_run_recipeA
  3.8/5.0
  Behavior3/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  Annotations are all false, so the description carries the burden. It discloses the dry_run behavior (no GUI operation) and the sequential nature plus the 100-action limit. However, it does not explain potential side effects of non-dry-run execution, failure handling, or what 'restricted' means in practice.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is one compact, front-loaded sentence covering core functionality and dry_run. No wasted words, every phrase adds value.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  This is a complex tool with an actions array of many fields, but the description omits crucial context: how to specify session_id, how actions are structured, error handling, and the meaning of 'evidence screenshots'. The output schema exists but does not replace the need for usage context.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 0%, so description must compensate. It explains the actions parameter by listing supported action types and describes dry_run explicitly. However, it does not mention session_id or how to structure the actions array, leaving significant gaps for parameter understanding.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly specifies the tool's function: sequentially executing up to 100 restricted GUI actions (menu, wait, click, type, key, sleep, capture). It also mentions dry_run validation, which distinguishes it from sibling tools that perform individual actions. This is a specific verb+resource statement.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description gives clear context: it is for executing a batch of actions in sequence, and dry_run mode is for validation only. It does not explicitly mention alternatives or exclusions, but the context strongly implies batch use versus calling sibling tools individually.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_set_stage_planA
  3.8/5.0
  Behavior4/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  The description adds the key behavioral constraint that after locking, stage requirements cannot be replaced or downgraded. Since annotations provide no safety hints (readOnly/destructive/idempotent all false), this disclosure is valuable. It does not cover whether the lock is reversible or what happens to an existing plan, but the core mutation and lock semantics are clear.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Two concise sentences deliver the core purpose and a critical constraint without extraneous details. The description is well-structured and front-loaded with the verb and resource.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  The tool has moderate complexity with 3 parameters and nested stage objects. The description covers the plan contents and lock behavior but omits session context and the meaning of require_output. An output schema exists, so return values are not needed. Overall, it is minimally viable but not fully complete for a planning/locking workflow.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 0%, so the description must compensate. It mentions stages, evidence types, required parameters, and expected windows, but omits session_id, require_output, caption, and action_summary. The description gives partial meaning to a few fields but leaves most parameters undocumented, especially the require_output flag.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool writes and locks an ordered stage plan with evidence types, required parameters, and expected windows. This specific verb and resource scope distinguishes it from siblings like run_recipe or capture_evidence, and the locking behavior is explicit.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies usage (present the guide-derived stages and lock them) but provides no explicit when-to-use guidance or alternatives. It doesn't state prerequisites (e.g., a live session) or when not to use it, so context must be inferred.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_capture_evidenceA
  3.9/5.0
  Behavior4/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  The description discloses key behaviors: capturing a specified window, saving PNGs sequentially, and writing specific fields (stage, caption, parameters, window title, time, SHA-256) to an evidence list. This goes beyond the annotations (which only set all hints to false) and provides useful side-effect context. It does not cover every edge case but adds meaningful behavioral information.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is a single sentence, action-first, with no filler. It efficiently conveys the tool's main actions and outputs while remaining concise and readable.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  With 7 parameters and no annotations, the description covers the core workflow (capture, save, log) but omits details about optional parameters like evidence_type and action_summary, and how session_id identifies the controlled window. Given the output schema exists, return values are already covered, but the description is not fully complete on its own.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, so the description must clarify parameters. It maps to stage_id (阶段), caption (说明), parameters, and window_title_contains (窗口标题), but leaves session_id, evidence_type, and action_summary unexplained. Time and SHA-256 are mentioned as outputs, not inputs. This partial compensation makes it adequate but not complete.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses the specific verb '截取' (capture) with a clear resource ('受控 ENVI 窗口') and enumerates concrete outputs (PNG files, evidence entries with SHA-256). This distinguishes it from sibling tools like finalize_evidence or list_windows, making the purpose unambiguous.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies usage for capturing step-by-step evidence, but it does not explicitly state when to use this tool versus alternatives (e.g., finalize_evidence) or mention prerequisites like an active session. It gives context but no exclusions or comparisons.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_invoke_menuA
  3.9/5.0
  Behavior3/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  Annotations are all false, providing no safety hints. The description adds that it uses native WM_COMMAND and does not depend on pixel coordinates, but it does not disclose requirements like an active session, potential dialogs, or error behavior. This is meaningful but incomplete for a tool that sends UI commands.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  A single, front-loaded sentence with one concrete example. Every word contributes meaning, and the key differentiator (not pixel-based) is included. No filler.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a simple invocation tool with an output schema, the description covers the core action and path format. However, it omits prerequisites like having a running session and does not mention that valid paths can be obtained via sibling tools like list_menu_items. Adequate but with evident gaps.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 0%, so the description must compensate. It explains menu_path well with a clear example showing an array of strings. However, session_id is not explained at all, and no syntax or format details are given beyond the example. Partial compensation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description states a specific verb and resource: 'invoke native WM_COMMAND' by full menu name path. It clearly distinguishes this from pixel-based clicking and from sibling tools like list_menu_items by focusing on invocation. The example path makes the purpose concrete.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The context is clear: use this to trigger a menu command by its full path. It doesn't explicitly reference alternatives or exclusions, but the provided example and the contrast with pixel-based methods give a clear sense of when this tool is appropriate. It lacks explicit 'use this when' wording but is more than merely implied.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_register_outputA
  4/5.0
  Behavior4/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  The description goes beyond annotations by stating that it does not modify result files and does not allow reading files outside the workspace. This is valuable safety context. However, it does not mention whether registration is idempotent or what happens on repeated calls, though idempotentHint=false is already provided.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is two short sentences, front-loaded with the primary action and immediately followed by important safety constraints. Every word earns its place; there is no verbosity or redundancy.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the tool has 4 parameters and an output schema, the description provides only the core action and two key constraints. It lacks parameter semantics and any process context (e.g., when to call relative to other tools), but the output schema and annotations fill some gaps. The description is minimally viable but leaves the agent to infer parameter usage.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, and the description provides no parameter-level details. The parameter names (path, label, metadata, session_id) are somewhat self-explanatory, but the tool description does not clarify constraints like path being workspace-relative or the format of metadata. The description fails to compensate for the low schema coverage.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool registers result files already generated in the experiment workspace and computes a hash. It uses a specific verb and resource, and distinguishes itself from the sibling 'envi_classic_register_inputs' by targeting outputs. This is a clear, specific purpose.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies usage for result files that have already been generated in the workspace, which differentiates it from input registration. However, it does not explicitly name alternatives or state when not to use it, so it stops short of full guidance.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_finalize_evidenceA
  4.1/5.0
  Behavior4/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  The annotations are all false, offering no safety or mutability hints. The description adds behavioral context by detailing the validation steps (checking stages, hashes, duplicates, order, output files) and the generation of JSON/Markdown packages, which implies a write operation. It also discloses the complete=false constraint, which is a meaningful behavioral rule. This goes beyond the minimal annotation information, but it does not mention side effects like overwriting existing files or failure behavior.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is concise (two sentences) and front-loaded with the primary action. Every clause contributes meaningful information: the checks, the output generation, and the critical condition. There is no redundancy or filler.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the tool's complexity (validation and generation) and the presence of an output schema, the description covers the key aspects: what is checked, what is produced, and a critical precondition. It does not explain error handling or what happens if validation fails, but the output schema may cover return values, and the description is otherwise adequate for a finalization tool.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  The schema has only one parameter, 'session_id', with no description (0% schema coverage). The tool description does not mention this parameter at all, failing to compensate for the low coverage. While the parameter name is self-explanatory, the description provides no additional meaning about its role or constraints, so the parameter semantics are weak.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool's function: verifying required stages, screenshot hashes, duplicates, order, and output files, then generating JSON and Markdown evidence packages. The verb '校验' (verify) and '生成' (generate) specify the actions, and the resource '证据包' (evidence package) is explicit. This distinguishes it from sibling tools like capture_evidence or register_output.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description provides a specific usage condition: 'complete=false 时不得进入报告生成' (must not enter report generation when complete=false), which guides when not to proceed. It also references '按指导书顺序' (per the instruction book order), implying a recommended sequence. However, it does not explicitly name alternative tools or exclude cases where other tools should be used, so it falls short of a perfect score.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_list_menu_itemsA
  4.2/5.0
  Behavior4/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  The description adds behavioral context beyond the read-only and idempotent annotations by disclosing pagination behavior and the ability to filter results with filter_text. It also notes that the menus are 'real' Win32 menus, which clarifies the data source. No contradiction with annotations.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is a single concise sentence that front-loads the primary action and purpose, followed by a useful note about filtering. There is no redundancy or unnecessary detail.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  The output schema likely covers return values, so the description need not explain them. The description covers pagination, filtering, and the resource type. It does not mention session requirements, but that is implied by the required session_id parameter and presence of sibling session-management tools.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, so the description compensates by explicitly explaining filter_text and implying limit/offset through '分页返回' (paginated return). Session_id is not described, but it is a common session context. This adds meaningful value over the bare schema.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool's function: it returns paginated real Win32 menu paths and command IDs from the ENVI Classic main window. This specific verb+resource combination distinguishes it from siblings like 'envi_classic_invoke_menu', which invokes menus, and 'envi_classic_list_windows', which lists windows.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies usage through pagination and filter_text but does not explicitly state when to use this tool versus alternatives or provide exclusions. It lacks direct references to sibling tools or conditions where this tool is preferred over others.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_register_inputsA
  4.2/5.0
  Behavior4/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  The description discloses key behaviors: per-file SHA-256 computation, writing to the evidence list, and the workspace constraint. Since annotations provide no hints (all false), this adds valuable context. However, it doesn't explain what happens on failure or whether re-registering overwrites entries, leaving some transparency gaps.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is a single, information-dense sentence. It front-loads the primary action and includes necessary constraints without redundancy, earning its place fully.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  The presence of an output schema means return values need not be described. The description covers purpose, timing, hashing behavior, evidence writing, and workspace constraint. It lacks session_id semantics and error scenarios, but overall it provides sufficient context for an agent to select and invoke the tool safely.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With schema description coverage at 0%, the description must compensate. It roughly maps guide_path, template_path, and data_paths to the guide, template, and data, but session_id is completely unexplained. The workspace constraint applies to all paths, adding some meaning, but session_id remains a gap and parameter details are shallow.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool's purpose: registering teacher-provided experiment inputs (data, guide, template) before screenshots, computing SHA-256 hashes, and writing them to the evidence list. The verb '登记' (register) combined with the explicit resource and action distinguishes it from sibling tools like register_output and finalize_evidence.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  It provides explicit temporal guidance ('在截图前' – before taking screenshots) and a hard precondition ('输入必须位于实验工作区内' – inputs must be inside the experiment workspace). It does not explicitly name alternatives, but the context of registering inputs vs. outputs is clear from the description and tool name.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_close_sessionA
  4.3/5.0
  Behavior5/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  Beyond the destructiveHint annotation, this description discloses essential behavior: graceful shutdown by default, retention on unsaved prompts, and force=true override. This enriches the raw annotation with operational nuance, which is exactly what the description should provide.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is two sentences, front-loads the core purpose, and every clause adds value. No redundant phrases or filler.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the tool's simplicity, annotations, and output schema, the description covers the primary flow and force mode well. The only gap is timeout_seconds behavior, which is a notable omission for a tool with only three parameters.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, so the description must compensate. It explicitly explains force=true's impact, but does not explain session_id or timeout_seconds. The timeout semantics are entirely missing, leaving a significant parameter unexplained.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states it closes only a specified ENVI process started by this service, using a specific verb and resource. It distinguishes from sibling tools like launch, health check, and menu invocation by emphasizing the 'only' scope and managed process.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies when to use this tool: when you need to close a managed ENVI session. The '只' (only) constraint sets boundaries, but there is no explicit alternative or when-not-to-use guidance. It provides a clear context without explicit exclusions.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

• envi_classic_health_checkA
  4.3/5.0
  Behavior4/5

  Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

  Annotations already declare readOnlyHint=true, destructiveHint=false, and idempotentHint=true. The description adds value by detailing what is read (registry and installation files) and the purpose (confirming safe startup), which goes beyond the annotations. However, it does not describe potential edge cases like missing files or permission issues, but that is not required given the read-only nature.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is a single, concise sentence that immediately states the action and target. It is front-loaded and contains no redundant information, making it easy for an agent to parse quickly.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness5/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the low complexity (zero parameters), rich annotations (read-only, idempotent), and the presence of an output schema, the description is complete. It specifies exactly what the tool examines (registry and installation files) and the goal (safe startup confirmation), which is sufficient for the agent to understand the tool's role in the workflow.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  The tool has zero parameters, and the schema has no properties. The description correctly focuses on the tool's behavior rather than parameter details. With no parameters to explain, the description fully covers the semantics needed for the agent to invoke the tool correctly.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool's purpose: reading the local registry and installation files to confirm whether ENVI Classic 5.6 (64-bit) can be safely started. The verb '读取' (read) and resource '注册表和安装文件' (registry and installation files) are specific, and the intent to verify safe startup distinguishes it from sibling tools like 'envi_classic_launch' which actually starts the application.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies the health check should be used before launching ENVI Classic, but it does not explicitly state when to use it or mention alternatives. No when-not-to-use conditions or exclusions are given, so guidance is limited to what the tool name and description suggest.

  Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.