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

With no annotations, the description carries the full burden and does well: it discloses the 50-file limit, that oversized directories are rejected before any conversion, that source .eml files are left in place (non-destructive), and output folder mirroring. It omits auth requirements, runtime/performance expectations, and what happens on partial failure beyond the 'failures' count.

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 short sentences, each earning its place: the operation and scope, the batch constraint plus output semantics, then the routing hint to convert_eml. Front-loaded 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?

The output schema exists, so listing the JSON summary fields is somewhat redundant but harmless, and the write/limit behavior is covered. The real gap is that a 15-parameter tool documents only one parameter, leaving the bulk of configurable conversion behavior unexplained.

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 description coverage is 0% across 15 parameters, so the description must compensate, yet it only explains output_directory semantics. The 13 remaining options (preset, dry_run, thread_mode/order, and eight strip/include toggles) get no explanation of defaults or effect, leaving the agent to guess at conversion behavior.

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?

States a specific verb and resource ('Batch convert all .eml files in a directory to Markdown') and immediately distinguishes itself from the sibling convert_eml by routing single-file retrieval there. An agent can tell exactly what this does versus convert_mbox or convert_eml_to_bundle from the first sentence.

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?

It names the alternative ('Use convert_eml to retrieve individual converted file content') and gives a clear batch context, plus the 50-file ceiling that defines when this tool is appropriate. It does not, however, explain when to choose this over convert_mbox or convert_eml_to_bundle, so it stops short of full when/when-not coverage.

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

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

With no annotations, the description carries the full burden and does well: it discloses that full Markdown is returned, that output_path additionally writes to disk, that attachments are listed but not decoded to disk, and the nuanced forwarding/quoted-text behavior in thread_mode. It omits error behavior and the exact effect of the unmentioned flags, which keeps it from a 5.

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?

Front-loads the core purpose, then delivers return behavior, the attachment caveat, thread-mode nuance, and a scannable preset list. Every sentence carries information and nothing is padded.

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?

An output schema exists, so the description needn't restate return values, and it appropriately focuses on behavior an agent can't infer. For a 14-parameter tool with zero schema documentation, the few flags left unexplained (thread_order, embed_inline_images, no_calendar_summary) are the only real gap.

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 description coverage is 0% across 14 parameters, so the description must compensate, and it does for the critical ones: it enumerates the four preset values, defines the preset-to-flag mapping, states that individual flags override the preset, and explains output_path and thread_mode. It leaves several booleans (embed_inline_images, no_calendar_summary, thread_order) unexplained, so compensation is strong but incomplete.

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?

States a specific verb and resource ('Convert a .eml email file to Markdown with YAML front matter') and immediately clarifies the output shape. It also names the sibling convert_eml_to_bundle for the attachment-saving case, so an agent can separate this tool from its neighbors without opening a schema.

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?

Gives clear routing advice for the attachment case ('use convert_eml_to_bundle to save the decoded attachment files') and explains when each preset applies and when thread_mode="structured" is appropriate. It stops short of contrasting with convert_mbox or convert_directory, which the sibling list shows are plausible alternatives.

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

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

With no annotations, the description carries the full burden and does meaningful work: it discloses that the original .eml is copied and left untouched, that move/delete are rejected over MCP, and that the result is a new directory of extracted artifacts. It does not cover overwrite behavior if bundle_root already exists, so it is not exhaustive.

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 purpose is front-loaded in the first sentence, followed by short, scannable paragraphs on output shape and the source_handling restriction. There is mild redundancy between 'self-contained bundle' and 'Creates a directory containing...' but nothing wasteful.

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?

An output schema exists, so enumerating return keys is largely redundant and the description need not explain return values. The real gap is the undocumented parameter surface: for a 15-parameter tool, an agent has no basis for setting the dozen optional flags.

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 description coverage is 0% across 15 parameters, so the description must compensate. It only explains source_handling (and even that partially) while leaving preset, thread_mode, thread_order, and the ten include_*/strip_* booleans entirely undocumented in both schema and prose.

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 states a specific verb and resource ('Convert a .eml file to a self-contained bundle') and clarifies the output artifact (a directory with markdown, attachments, and optionally the source). It is distinguishable from convert_eml by the 'bundle' framing, but it never explicitly contrasts itself with that sibling, so it stops short of 5.

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?

It gives a real usage constraint for one parameter: source_handling only accepts 'copy' over MCP, and move/delete must be done via CLI or Python API. However, it offers no guidance on when to choose this tool over convert_eml, convert_mbox, or convert_directory, so tool-level selection is left to inference.

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

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

With no annotations, the description carries the full burden and does so well: it discloses size and message-count bounds, truncation signalling (truncated=true), rejection cases, the non-destructive guarantee ('The source archive is never modified'), collision-safe report output, and the absence of cancellation. These are exactly the operational traits an agent needs before committing to a long-running conversion.

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?

Front-loaded with the core action, then organized into labelled blocks ('Bounded:', output behavior, 'Returns a JSON summary'). Dense but every sentence adds distinct operational information; nothing is 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 16-parameter tool with no annotations, it covers purpose, bounds, failure handling, side effects, and return shape (with the output schema handling the rest). An agent has enough to call it correctly and knows the edge cases that cause rejection or truncation.

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 0% across 16 parameters, yet the description only explains two of them (output_directory is required, bundles=true switches to bundle directories). The remaining toggles (preset, thread_mode, strip_*, embed_inline_images) are undocumented, though most have self-describing names. Partial compensation only.

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 states a specific verb and resource ('Convert one flat .mbox archive ... to Markdown files') and immediately narrows scope with 'flat' plus rejection of compressed archives and Apple Mail .mbox directories. This lets an agent distinguish it from convert_eml and convert_directory without opening another schema.

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?

It gives concrete when-to-use conditions (archive must be ≤256 MiB, ≤1000 messages) and an explicit alternative path ('use the dead-letter CLI for larger archives'). It does not directly route between convert_mbox, convert_eml, and convert_directory, but the format constraints make the boundary largely inferable.

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

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

With no annotations, the description carries the full burden and does reasonable work: it discloses the non-destructive trait ('without writing permanent files'), the always-JSON return, the state enum values (normal/degraded/review_recommended), and that some keys are conditional. It omits error behavior and any cost/latency characteristics, but the core behavioral profile is present.

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?

Purpose and usage are front-loaded in two compact sentences, which is good. The closing enumeration of return keys largely duplicates the existing output schema and consumes several lines without adding routing or setup value.

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?

Return values are covered by the output schema, so the description's key list is redundant rather than additive. For a 13-parameter tool with zero schema coverage, the description leaves the whole configuration surface undocumented, which is a substantial completeness gap despite the adequate purpose/usage framing.

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?

There are 13 parameters at 0% schema description coverage, so the description is expected to compensate and instead says nothing about any of them. Preset, thread_mode, thread_order, and the five-plus strip_* toggles are entirely unexplained, leaving the agent no basis for setting them.

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 states a specific verb and resource ('Inspect email quality and structure') and adds a differentiating constraint versus the convert_* siblings: it operates 'without writing permanent files'. That implicitly separates it from the conversion tools, though no sibling is named explicitly.

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?

It gives clear trigger conditions — 'assess conversion quality before committing' and 'troubleshoot problematic .eml files' — which tells the agent when this inspection step is warranted. It stops short of naming an alternative tool or stating exclusions.

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