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

With no annotations, the description carries the full burden. It discloses one genuinely useful behavior (search/get read the index as of `as_of`) and that `add` registers a hand-downloaded file, but `index` and `fetch` are dismissed as "tasks on the corpus resource" with no mutation, async, or permission details.

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 short and front-loads the action list, but the final sentence ("`index` and `fetch` are tasks on the corpus resource") consumes space without conveying meaning. Some structure exists, but the economy comes partly from under-specification rather than 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?

A 12-parameter multi-action tool with mutating actions and no annotations needs substantial behavioral context. The presence of an output schema excuses return-value documentation, but action-to-parameter mapping and mutation semantics are largely missing.

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 12 parameters, so the description must compensate. It touches only `action`, `as_of`, and (implicitly) `key`; the meaning and per-action relevance of `k`, `id`, `ids`, `url`, `file`, `kind`, `query`, `region`, and `factors` is left 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?

It names the resource ("scenario corpus") and enumerates the six actions, which is more than a tautology, but it never says what a scenario corpus is or how it differs from the sibling `scenario` or `series` tools. An agent still has to infer the tool's domain role.

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?

Per-action hints exist (search/get read the index, list shows the manifest, add registers a file), so action selection is partially implied. But there is no explicit when-to-use versus the `scenario`, `series`, or `run` siblings, and no statement of preconditions or when not to call it.

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?

No annotations are provided, so the description carries the full behavioral burden. The action names hint that 'cancel' may mutate state and 'log'/'status' are reads, but it does not disclose side effects, permissions, async behavior, or whether cancellation is reversible.

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 very short, front-loaded with the domain ('Workspace and task state'), and free of filler. It is concise, though the extreme brevity leaves much unsaid.

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?

This is a multiplexed action tool with three parameters, no annotations, and 0% schema description coverage. The description lists action names but omits parameter meanings and action-specific behavior; the presence of an output schema only relieves the need to explain return values.

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 must compensate. It partially helps by listing possible action values ('state, status, log, cancel'), but it does not explain the 'id' or 'tail' parameters at all.

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 says 'Workspace and task state' and lists four action names, giving a rough idea that this is a multiplexed jobs/task tool. However, it does not state a specific verb-resource pair for each action or distinguish the tool from siblings like run or scenario.

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 about when to use this tool versus alternatives such as run, scenario, or series. The action list names operations but provides no context, prerequisites, or exclusions.

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 of behavioral disclosure, yet it never says whether estimate and backtest are heavy compute jobs, whether they are read-only or produce persisted state, how long they take, or whether they require credentials. The version-mapping sentence is structural, not behavioral.

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 resource, kind values, and action list, then a single sentence tying version to action. It is tight and free of filler, though the terse 'kind: macro, satellites' phrasing reads as shorthand rather than a sentence.

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 return values need not be described, and the description is proportionate for a compact multi-action dispatcher. The only genuine gap is documentation of 'only' and 'as_of', which is already penalized under parameter semantics rather than counted again here.

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 must compensate for all five parameters. It clarifies the accepted values for 'kind' (macro, satellites), 'action' (list, describe, estimate, backtest) and the per-action meaning of 'version', but 'only' and 'as_of' are never mentioned at all, leaving two parameters completely undocumented anywhere.

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 the resource ('Model versions') and enumerates two kinds and four actions, so an agent can tell this is a dispatcher tool over model artifacts. However, no verb phrase explains what 'describe', 'estimate' or 'backtest' actually do or return, leaving the purpose only partly resolved.

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 only routing guidance is the negative rule that 'list takes no version.' There is no statement of when to choose estimate versus backtest, no prerequisites, and no differentiation from siblings such as series, scenario, or run.

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 it does disclose useful traits: `save` validates a contract and writes to `portfolios/<id>.json`, `check` reports unservable positions with rejection reasons, `exposure` values at the scenario's macro calculation month. However, `delete`'s destructiveness/irreversibility and permission requirements are never addressed.

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 action list, which is the right structure, and it is short. But the compression goes too far — telegraphic clauses and inconsistent backtick markup make it denser than it needs to be, and half the actions get no sentence at all.

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?

A six-action dispatcher with no annotations and 0% parameter coverage explains only three actions, says nothing about `delete` semantics, and never clarifies which parameters apply to which action. The presence of an output schema covers return values, but the call-shape gaps remain significant.

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 four parameters, so the description must compensate and largely doesn't. The `<id>.json` path hints that `id` is a filename component and `save`'s contract check hints at `spec`, but `scenario` and `action` values are only implicitly defined, and usage per action (which params each action needs) is never stated.

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 the resource (stored portfolios) and enumerates all six valid actions, which is essential because the `action` parameter has no enum. It distinguishes itself from siblings like `scenario` and `run`, but leaves the domain meaning of a 'portfolio' and the shape of a 'spec' unstated.

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 text describes what `save`, `check`, and `exposure` do, implying when to reach for them (writing, validating coverage, valuing), but never states when to use `list`/`get`/`delete` or how to choose between them. It gives no exclusions and no pointer to a sibling tool that might be a better fit.

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 disclosure burden. It does add one genuinely useful behavioral fact: `start` is asynchronous and materializes 'a job on the ``runs`` resource', so the agent knows to poll `status` rather than expect synchronous results. Beyond that it is silent on cost, permissions, side effects, and whether repeated starts are idempotent — notable gaps for a tool whose `start` action triggers computation.

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 action enumeration, then a dense explanatory paragraph — good ordering for an agent scanning for the entry point. The final sentence about `base_currency` and portfolios is slightly elliptical ('a portfolio brings its own'), but overall there is little 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?

This is an 8-parameter, four-action tool with zero schema descriptions, and the description leaves the parameter-to-action mapping implicit and four parameters entirely undocumented. The output schema existing means return values need no explanation, which helps. But an agent still has to guess which fields `status`/`result` need versus `start`, so the definition is only adequate, 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 description coverage is 0%, so the description must compensate, and it only partially does. It usefully defines `base_currency` as the base of `positions` and clarifies that a portfolio supplies its own base currency, and it enumerates the legal values of `action` (critical, since the schema types `action` as an unconstrained string with no enum). It says nothing about `id`, `seed`, `draws`, or `scenario`, nor which parameters apply to which action.

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 gives a concrete verb-plus-resource framing ('Runs on the paths of a scenario') and then enumerates the four actions with a one-line explanation of each (start draws paths and computes positions, status reports the job, result returns per-position results, list enumerates workspace runs). An agent can tell it is the execution/read-back surface for scenarios, distinct from siblings like `scenario` or `portfolio`. It stops short of explicitly naming how it differs from the `jobs` sibling, which holds it at 4.

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 action-level breakdown effectively functions as when-to-use guidance: use `start` to launch a run, `status` to poll the resulting job, `result` to pull outputs, `list` to enumerate. That covers the main decision points an agent faces with a multi-action tool. There is no exclusion guidance (e.g., when a run should be done via a sibling tool, or that `status`/`result` require an `id` from `start`), so it is not a 5.

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 add real signal: save has a filesystem side effect at scenarios/<id>.json and performs contract checking, while validate reports every blocking problem including corpus fragments not visible on the calculation day. However, list and get are completely undocumented, and there is no mention of overwrite semantics, permissions, or whether save is destructive.

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 action list is front-loaded in the first sentence, which is the right structure for a multiplexed tool. The second sentence is long and clause-heavy ('including corpus fragments not visible on the calculation day when the index exists') but the content is substantive rather than 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?

An output schema exists, so return values need not be described. Still, for a four-action tool with zero schema description coverage, the description omits the per-action parameter mapping and never says what list or get do, leaving meaningful gaps an agent must guess at.

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 'action' is a bare string, so enumerating the four valid action values is genuinely necessary and the description supplies it. But 'spec' is never explained anywhere and 'id' is only implied through the file path pattern; which parameters apply to which action is left to inference.

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 the resource (stored scenarios) and enumerates the four concrete actions (list, get, save, validate), so an agent knows what the tool manipulates. It does not differentiate itself from siblings like 'series', 'portfolio' or 'model', which likely also deal with persisted objects, so an agent cannot route purely on purpose text.

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 implicitly distinguishes save (writes a file after checking the contract) from validate (diagnoses why a scenario cannot be computed), which is a useful internal alternative. But there is no statement of when to use this tool instead of 'series', 'run' or 'portfolio', and no prerequisites or when-not conditions.

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?

No annotations are provided, so the description carries the full burden of behavioral disclosure. It explains a non-obvious update mode (fetch by ids versus discover by quantity), but it says nothing about whether add/update mutate data, permission requirements, validation, or what search/get return.

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 short sentences, front-loaded with the resource and action list, with no filler. Its terseness is a completeness issue rather than a conciseness issue.

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 seven-parameter, four-action tool with no annotations and no schema parameter descriptions, the description covers only one branch of one action. The output schema may cover return values, but input semantics and most action behavior remain largely undocumented.

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 must compensate for seven undocumented parameters. It only gives meaning to ids and quantity in the update-specific note, leaving id, text, as_of, entry, and the action values 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?

The description names the resource ('Stored series') and enumerates the four actions, so an agent can identify this as the series-management tool rather than corpus, scenario, run, model, portfolio, or jobs. It is clear at the resource/action level, though it does not explain what search, get, or add actually do.

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 does not say when to use search versus get versus add versus update, nor when to prefer this tool over its siblings. The only conditional guidance is the internal update behavior involving ids and quantity.

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