Seiche Market Atlas Corpus

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

Annotations already declare readOnly, idempotent, non-destructive, and closed-world, so the safety profile is covered. The description adds genuine context beyond that: it reveals the tool returns an inventory of content-addressed shards and that binary payloads are NOT included and must be retrieved via REST — a real behavioral constraint that shapes how an agent chains calls.

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?

One compact sentence with the core resource front-loaded and the important caveat trailing after a semicolon. Every clause carries information; nothing is redundant.

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 one-parameter, non-destructive read with no output schema, the description covers what is produced (shard inventory) and the key limitation (no binaries). It falls short on the one thing left undocumented: what flow_id is and where to obtain it, which is essential to actually invoking the tool.

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 is a single required parameter, flow_id, with 0% schema description coverage and no explanation in the description. An agent can only infer that flow_id scopes the inventory to a BIS flow; format, expected values, and how it relates to the sibling tools' identifiers are undocumented.

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 ('Discover') and a specific resource ('content-addressed BIS shard inventory'), which distinguishes it from the data-returning siblings bis_observations and bis_records. It does not name a sibling or explicitly contrast itself, and the jargon ('content-addressed shard inventory') assumes domain knowledge.

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 clause 'binary shards remain REST downloads' implies a when-not: do not use this tool to fetch shard binaries, follow up over REST. However, there is no explicit guidance on when to choose this manifest over bis_observations, bis_records, or corpus_catalog, so selection between siblings 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?

Annotations already declare read-only, idempotent, closed-world, non-destructive behavior, so the safety profile is fully covered. The description adds 'bounded' (implying the limit cap) and 'revision-aware', but the latter is never explained and there is no revision parameter in the schema, leaving the agent guessing what revision-awareness implies for results.

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?

A single short phrase with no filler, so it is not bloated. But it is under-specified rather than concise—terseness here costs clarity rather than saving an agent anything.

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?

No output schema, no annotations gaps covered by prose, and undocumented parameters leave the definition thin. For a data-retrieval tool with a required flow_id and pagination-like limit, the description should at minimum say what is returned and how the flow scopes results.

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?

With 0% schema description coverage and two parameters, the description must carry the load: it never mentions flow_id as a required flow selector, nor explains limit semantics beyond the vague word 'bounded'. The agent must infer both parameters from names alone.

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?

Names a verb (read) and resource (BIS observations) with qualifiers 'bounded' and 'revision-aware', so the basic operation is inferable. But it offers no differentiation from close siblings like bis_records or seiche_observations, and 'revision-aware' is undefined jargon that doesn't help the agent distinguish use cases.

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 gives no guidance on when to choose this over bis_records, bis_flow_manifest, or seiche_observations. There are no stated preconditions, exclusions, or alternative routes despite the crowded sibling set.

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?

Annotations already declare read-only, idempotent, non-destructive, closed-world behavior. The description adds that results are 'bounded' and that complete transfer should use the manifest/shard path, but it does not explain pagination behavior or what bounded 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.

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

The description is a single front-loaded sentence with no wasted text. It is concise and immediately communicates the browsing scope plus the complete-transfer alternative.

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?

With no output schema and no parameter descriptions, the definition leaves important operational details unspecified: what BIS records contain, how pagination via cursor works, and how limit affects results. The manifest alternative helps, but the tool is still under-described for its three-parameter interface.

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%, so the description needs to compensate for undocumented parameters. It weakly implies bounded results through 'bounded' but does not explain flow_id, limit, or cursor semantics.

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: browsing bounded BIS records. It also hints at the distinction from complete transfer via the flow manifest and shard URLs, though the domain-specific BIS terminology and 'full-fidelity' wording remain somewhat opaque.

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 clear use case for browsing bounded records and explicitly points to the flow manifest and shard URLs for complete transfer. That provides a meaningful alternative path, though it does not fully contrast with siblings like bis_observations.

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?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile needs no repetition. The description adds only the breadth of the scan (all corpora plus evidence classes); it says nothing about cost, size of the result, or whether the summary is aggregated versus per-corpus.

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?

A single seven-word sentence with the scope qualifier ("every registered") front-loaded and no filler. Slightly terse: "evidence class" is domain jargon left undefined, which costs a bit of clarity without adding bulk.

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?

With no output schema and no parameters, the description is the only place an agent can learn what the summary actually contains, and it does not say — counts, identifiers, or prose per corpus are all unspecified. An agent can decide to call it, but cannot predict the shape of the response before doing so.

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 tool takes zero parameters and the schema is closed (additionalProperties=false), so there is nothing for the description to disambiguate. Baseline 4 applies: no parameter semantics are needed and none are wrongly implied.

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?

Specific verb ("Summarize") plus two named resources ("every registered corpus and evidence class"), which is more than a restatement of the name. It does not, however, distinguish itself from the adjacent corpus_health or list_datasets, so an agent must guess whether this overlaps them.

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 word "every" implies a catalog-wide overview rather than a targeted lookup, which weakly separates it from list_datasets and corpus_health. There is no explicit when-to-use, no exclusion, and no named alternative, so the routing decision 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?

The annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is fully covered structurally. The description adds only the useful detail that the check covers both the service and the storage roots, but says nothing about what a failure looks like or what the response contains.

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?

A single six-word phrase with zero filler, and the resource being checked is front-loaded. It is arguably too terse for a health probe, but nothing in it is wasted.

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?

With no parameters, no output schema and rich annotations, the definition is close to sufficient, but because there is no output schema the description should say what a health result looks like (status object, root list, error behavior). That gap keeps it at adequate rather than 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?

The tool takes zero parameters and the schema is a closed empty object, so there is nothing for the description to disambiguate. Baseline 4 applies.

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 concrete verb and resource: it checks the corpus service and storage roots, which is a health/diagnostic operation clearly distinct from data-querying siblings like corpus_catalog or list_datasets. The only shortfall is that 'Check' doesn't say what is verified or returned, but the scope is specific enough that an agent can tell what the tool is for.

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?

There is no guidance on when to call this tool versus alternatives, no prerequisites, and no mention of failure conditions or when it should be avoided. The agent must infer that this is a health probe and that no siblings cover that need.

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?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds one genuinely useful behavioral fact — that it does not read raw storage — but says nothing about what an invalid dataset_id does or what the inspection yields.

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?

A single front-loaded sentence with zero filler and the key constraint ('without reading raw storage') stated up front. Its brevity borders on under-specification, but there is no wasted text.

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 one-parameter read tool with annotations covering safety, an agent still lacks the essentials: what 'inspect' returns, where dataset_id comes from, and how this differs from corpus_catalog/corpus_health. With no output schema, the description should carry that burden and does not.

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% and the single dataset_id parameter is undocumented in the schema. The description only implies it takes 'a dataset' ('canonical public-index dataset'), giving no format, no provenance (e.g., from list_datasets), and no constraints, so it does not compensate for the coverage gap.

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?

It states a specific verb+resource ('inspect ... dataset') and implies a singleton counterpart to list_datasets. However, 'inspect' is vague about what actually happens — metadata summary? schema? sample rows? — and it does not distinguish itself from corpus_catalog, corpus_health, or the bis_/seiche_ observation tools that presumably also surface dataset-level information.

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?

There is no explicit when-to-use or when-not-to-use guidance, and no alternatives are named. The only implied usage cue is the singular 'one ... dataset' versus list_datasets' plural, which an agent must infer on its own.

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?

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered structurally. The description adds only that datasets are 'versioned' and carry rights/eligibility attributes, saying nothing about pagination behavior, default page size against the limit of 1000, or what the listing includes.

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?

A single front-loaded sentence with zero filler text. It is efficient, though the brevity comes at the cost of coverage rather than from disciplined editing.

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 8 optional filters, no output schema, no enum hints and 0% schema documentation, the description is far too thin to guide invocation. Eight sibling tools in the namespace make the missing scoping context more costly.

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 8 parameters, so the description carries the full burden and mostly fails: 'rights' and 'product eligibility' hint loosely at acquisition_review or data_class but never map to any named filter, and group, engine, collection_kind, publication_state, limit and cursor go entirely unexplained.

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 (List) and resource (versioned datasets) plus the scope of the listing (rights and product eligibility), which is more than a restated name. However, it gives no hint at how this differs from the sibling corpus_catalog or inspect_dataset, so the agent cannot route on purpose alone.

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?

There is no when-to-use guidance, no statement that this is a broad enumeration versus a filtered lookup, and no mention of the sibling tools (corpus_catalog, inspect_dataset) that likely overlap. The only usable signal is the verb 'List'.

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?

Annotations already declare readOnlyHint, idempotentHint and closed-world, so the safety profile is fully covered by structured data. The description adds only the domain note that the coverage is 'normalized' Seiche money-market data; it says nothing about result size, ordering, or whether the list is static or reflects live availability.

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?

A single five-word phrase, front-loaded with the verb. Nothing is wasted, but the extreme terseness edges into under-specification rather than economical precision.

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?

With no output schema, the description must convey what a caller gets back, and 'coverage' does not. For a zero-param enumeration tool whose entire value is the returned list, failing to describe the shape or granularity of that list leaves a 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?

The tool takes zero parameters with 100% schema coverage, so there is nothing for the description to disambiguate. Baseline 4 applies — no semantic gaps are possible with an empty input schema.

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 verb 'List' and the domain qualifier 'Seiche money-market' are specific, but the object 'coverage' is undefined — an agent cannot tell whether this returns markets, tickers, date ranges, or dataset availability. It is distinguishable from seiche_observations only by the ambiguous word 'coverage', not by an explicit statement of difference.

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?

There is no indication of when to call this rather than seiche_observations, list_datasets, or corpus_catalog — all plausible siblings for discovering what data exists. No prerequisites, no exclusions, no routing guidance of any kind.

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?

Annotations already cover read-only, idempotent, non-destructive, closed-world behavior, so the bar is lower. The description adds two real facts beyond them: results are 'bounded' (paginated/capped) and are only served after a source-wide redistribution validation, which is meaningful freshness/correctness context. It does not explain what happens if validation has not run, or what 'canonical' excludes.

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?

A single front-loaded sentence with no filler and no repetition of structured fields. It is efficient, though arguably undersized for an 11-parameter tool, which the completeness score penalizes separately.

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 an 11-parameter, no-output-schema tool with 9% description coverage, this is far too thin. Terms like 'canonical' and 'redistribution validation' are unexplained jargon, and nothing tells an agent how the enum filters (quality, staleness) interact or what a row contains.

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 only 9% across 11 parameters, so the description carries the burden and largely fails: 'bounded' weakly gestures at limit/cursor and 'source-wide' at source, but market, quality, staleness, event_from/event_to, knowledge_at, instrument_id, and semantic_role are undocumented in both schema and text.

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 ('Read') and resource ('canonical Seiche rows') with a scope qualifier ('bounded') and a stated precondition ('after source-wide redistribution validation'). It is distinguishable from seiche_markets, but it never names a sibling or clarifies how it differs from the other *_observations / *_records tools.

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 phrase 'after source-wide redistribution validation' implies a precondition but never states when an agent should choose this tool over seiche_markets or the bis_* observation tools, nor any exclusions. Usage must be inferred from the tool name alone.

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