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. It does disclose that the operation clears needs_attention, but it omits permissions, reversibility, error behavior for unfinished runs, and the meaning of cross_bot.

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 wording. The parenthetical efficiently explains the key state change.

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 mutation tool with no annotations and partial schema coverage, the description is too sparse. The output schema reduces the need to explain return values, but key invocation context around run selection and cross_bot behavior is 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 only 33%, and the description adds little parameter meaning. 'Finished runs' loosely implies run_ids, but cross_bot is not mentioned at all, and run_ids constraints are left entirely to the 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 description gives a specific verb and resource: 'Mark finished runs as seen,' and clarifies the effect as clearing needs_attention. It is clear what the tool does, though it does not explicitly distinguish itself from siblings such as session_status or session_tail.

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?

'Finished runs' implies the condition for use, but the description does not state when to use this versus alternatives like session_status, nor does it mention exclusions such as active runs. Usage is implied rather than explicitly guided.

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 one genuinely surprising trait: the call reconciles state and starts queued runs, i.e. a list operation with a mutation side effect. However, it is ambiguous whether this happens unconditionally or only under some flag, and nothing is said about permissions, whether queued runs can fail or block, or whether the call is safe to repeat.

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, front-loaded with the core action and ordering, then the parameter note, then the side effect. Nothing is padded, though the side-effect sentence reads as a tacked-on afterthought rather than being flagged as important.

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 explained, and the core action is covered. But for an eight-parameter tool with zero annotations and a hidden state-mutating side effect, the description leaves too much unstated: most filters are undefined and the conditions for reconciliation/queue-starting are unclear.

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 only 13% (just 'since'), so the description must compensate and does so for only one of eight parameters: include_external. bot, view, limit, status, cwd_prefix, and needs_attention are left entirely to bare titles, and the enum values for status/view are never explained.

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 ('List runs') plus ordering ('newest first') and that totals are included, which is more than the bare name conveys. It does not explicitly contrast with siblings like session_status or session_tail, but the listing scope is unmistakable.

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 statement of when to use this rather than session_status, session_tail, or session_result, which are the obvious alternatives for inspecting sessions. The only guidance offered is about a single parameter (include_external), not about tool selection.

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 useful work: it discloses that output is bounded and that answer text is wrapped under `untrusted`, a meaningful safety signal. However, it says nothing about whether the call blocks/waits for an unfinished run, what error occurs for an unknown or still-running run_id, or whether retrieval is read-only and side-effect free.

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 compact sentence that front-loads the core purpose and packs the payload inventory into a parenthetical with no filler. Minor cost: terms like 'worktree diffstat' and the `untrusted` wrapper convention are used without any gloss.

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 is right not to enumerate return values in detail, and it correctly flags boundedness and the untrusted payload. The remaining gaps — behavior on an in-progress run, run_id format, and the include selector — are substantial for a tool with three undocumented parameters and zero annotations.

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: '(bounded)' hints at max_chars truncation and the listed return fields loosely map to the `include` options. It never explains run_id semantics, the five `include` enum values (result, structured, denials, diffstat, stderr_tail), their default-null meaning, or the 100–20000 range and 2000 default on max_chars.

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 precisely what the tool returns: the final result of a run, enumerating status, cost, turns, worktree diffstat, and answer text. It is clearly distinct from session_start/session_send/session_stop by being a retrieval of terminal run output, but it never contrasts itself with session_status or session_tail, so an agent must infer which accessor to use.

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 statement. 'Final result' weakly implies a completed run, but the description never says what to call while a run is still going (session_status/session_tail) or what happens if the run hasn't finished, leaving the primary selection decision 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; it discloses the busy-contention policy (reject default, queue, fork) and the return shape (new run_id and wait_cmd). It omits whether the call mutates state, cost/permission implications, and cross-bot auth requirements — meaningful gaps for a mutation-style tool with 18 params.

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?

Two tight, front-loaded sentences with no filler; the busy-handling behavior and return values come first. It is efficient but arguably too terse for an 18-parameter tool.

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 18-param tool with no annotations and only 28% schema coverage, the description is thin — it covers busy handling and return values (aided by the output schema) but leaves most parameters, cost behavior, and mode semantics undocumented. An agent would need to guess at much of the surface.

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 only 28%, so most of the 18 parameters (mode, model, effort, notify, max_turns, timeout_min, allowed_tools, cross_bot, continue_interrupted, run_id) are undocumented in both schema and description. The description only illuminates if_busy/fork semantics, leaving the bulk of the parameter surface 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+resource: 'Follow up on an existing session (claude --resume)', which clearly contrasts with session_start's new-session role. The parenthetical anchor makes the operation unambiguous, though it stops short of naming the sibling 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?

'Follow up on an existing session' implies this is for continuing a session already in flight, and the busy-handling sentence gives conditional guidance for reject/queue/fork. However it never explicitly says when to use this vs session_start or session_wait, and there are no 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?

No annotations are provided, so the description carries the full behavioral burden. It usefully discloses that this is an asynchronous run returning run_id plus a wait_cmd to run in a background shell, and that edit mode creates a worktree unless isolation='none'. It omits other traits an agent needs: what happens when a run is busy but queue_if_busy is false, cost/budget enforcement behavior, concurrency limits, and side effects of edit mode on the working tree.

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 tight sentences with zero filler: the action, the return contract, and the safety-mode caveat are all front-loaded. Every clause carries information an agent needs before invoking.

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 22-parameter, no-annotation, long-running job tool, the description is too thin. Output schema coverage excuses explaining return shapes, but the description still leaves queueing behavior, cost control, concurrency, and most parameter meanings undocumented, so an agent cannot configure a run confidently from this definition alone.

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 only 27% across 22 parameters, so the description must compensate and largely does not. It only illuminates mode and isolation; the other high-impact parameters (priority, notify, notify_ref, trust, effort, max_turns, allowed_tools, timeout_min, idle_timeout_min, fallback_model, json_schema, name_slug, chat_hint, queue_if_busy) get no explanation in either the schema or the description.

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 ('Start a new headless Claude Code run') plus the immediate consequence ('Returns run_id and a wait_cmd'). The 'start' verb is unambiguously distinct from the sibling set (ack/status/list/tail/result/send/stop/wait), so an agent can route to it without opening the 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?

The description conveys mode semantics (default read-only; edit mode uses a git worktree unless isolation='none') and that runs can be queued, which implies usage. But it never says when to choose session_start over session_send for dispatching work, nor any prerequisites or exclusions. Usage is implied rather than stated.

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 burden alone; it does disclose the shape of the payload and flags the last text field as untrusted, which is genuinely useful security context. It does not state that the call is read-only, whether it fails for finished runs, or what run_id must reference.

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 compact sentence with the key payload fields front-loaded and no filler. The parenthetical '(untrusted)' is terse but earns its place.

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?

Because an output schema exists, the description needn't explain return values, and it doesn't. What remains missing is operational context for a no-annotation tool: read-only nature, lifecycle constraints, and how run_id is sourced relative to the sibling session tools.

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 single parameter run_id has 0% schema description coverage, and the description never mentions it — no format, no guidance on obtaining it from session_start. A one-line note would have compensated entirely.

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 specific resource (one run) and enumerates what the status contains (counters, tokens, last tool, budget, last text), which lets an agent distinguish it from session_list or session_tail. It lacks an explicit verb and never names a sibling for contrast, keeping it short of a 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?

There is no statement of when to call this versus session_tail, session_result, or session_list, all of which plausibly overlap. The only implicit signal is 'one run', which does not tell an agent which of several observability tools to pick.

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; it does valuable work by spelling out the SIGINT→SIGTERM→SIGKILL escalation for graceful mode. However it says nothing about what force mode does, whether queue removal discards a pending run irrecoverably, what cross_bot implies for authorization, or whether the operation 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?

A single dense sentence with the escalation detail parenthesized so it doesn't interrupt the main clause. Zero wasted words, and the primary action is front-loaded.

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 needn't be described, but for a five-parameter destructive operation with no annotations and 20% schema coverage, the description leaves too much unspecified — force mode, cross_bot semantics, and the fate of a removed queued run are all unaddressed.

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 only 20% — just `bot` is documented — and the description explains none of the other four parameters. It implies the meaning of mode=graceful via the signal sequence but never mentions the `force` enum value, `reason`, `cross_bot`, or what `run_id` refers to for a queued run.

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 specific verb and resource ("Stop a run ... or remove it from the queue") and even discloses the escalation sequence for graceful stops. It is clearly distinct from the session_* siblings (start/ack/status/list/tail/result/send/wait), none of which terminate anything, though it doesn't explicitly contrast itself with 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?

There is no statement of when to use this versus alternatives, no prerequisites (e.g. does stopping a queued run differ from a running one?), and no guidance on when to choose mode=force over the default graceful path. The agent must infer the conditions from the enum alone.

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 must carry behavioral context. It usefully discloses that events are condensed to ≤300 characters, are untrusted, and support pagination via next_seq, but it omits read-only/mutation behavior, ordering, permissions, and other operational traits.

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 tightly written sentences with no wasted words, and the core purpose plus pagination instruction are front-loaded.

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?

Although an output schema exists and reduces the need to explain return values, the tool has five parameters, no annotations, and 0% schema description coverage. The description is too thin to fully equip an agent to choose and invoke this tool correctly.

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

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

Schema description coverage is 0%, so the description must compensate for five undocumented parameters. It clarifies since_seq pagination and implies character truncation, but does not explain run_id, kinds, max_chars, or max_items, leaving most parameters ambiguous.

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 clearly identifies the resource being returned: condensed recent events for a specific run. It does not name an explicit retrieval verb or distinguish this from siblings like session_result or session_status, but the core purpose is still understandable.

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 usage instruction is pagination mechanics: 'Pass next_seq back as since_seq.' It does not say when to use this tool versus session_result, session_status, or session_wait, nor does it state any exclusions or prerequisites.

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 does the important work of disclosing that this call blocks (long-poll) and is capped at ≤50 s, which prevents an agent from assuming it is a cheap non-blocking read. It does not say what happens on timeout expiry or whether the call is side-effect free.

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?

Two short sentences, zero filler, with the blocking cap stated first and the alternative deferred to the end. Nothing to trim.

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 core blocking contract is covered. What remains missing for a tool with no annotations is the timeout-expiry behavior and confirmation that the poll has no side effects.

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 carries the burden and it does explain the `until` semantics (terminal vs. needs attention) and the 50 s ceiling that bounds `timeout_s`. It never mentions the 30 s default or what `timeout_s` does on expiry, leaving a real gap for a 3-parameter tool with no schema 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 names a specific operation (a bounded long-poll) on a specific resource (the run) and states the terminating condition (terminal or needs attention). It is clearly distinguishable from a status snapshot, though it never explicitly contrasts with the closely related session_status sibling.

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?

"For long runs use the Shell wait_cmd instead" gives an explicit alternative plus the condition that selects it, which is exactly the kind of routing guidance an agent needs. It stops short of saying when *not* to block at all (e.g. use session_status for a non-blocking peek).

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