humanize_ambient

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

With no annotations supplied, the description carries the full burden and does so well: it discloses the yield-to-every-verb precedence rule, the horizon_s timeout and idle-freeze termination, and that it never clicks/types/selects/navigates. It also volunteers an honest limitation — targets are best-effort because script-attached hover handlers are invisible, so a page may still observe a mouseover. It omits what 'status' reports and whether the setting survives restarts, which keeps it off 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?

The core opt-in/default-off constraint and the never-clicks guarantee are front-loaded, and the parenthetical compaction packs a lot of behavioral detail without padding. It reads as a dense run-on with internal shorthand ('yields to every verb', 'horizon_s', 'idle-freeze') that costs some immediate parseability, but no sentence 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?

For a zero-required-parameter toggle with no annotations and no output schema, the behavior is covered thoroughly, including edge cases and caveats. The gaps are contextual: nothing explains the relationship to the three sibling humanize_* tools, nor what a status call returns, which matters because mode defaults to status and no output schema describes the result.

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 100%, so all five parameters are already documented, and the baseline is 3. The description adds modest behavioral meaning by tying horizon_s to 'stops horizon_s after the last verb' and by framing scroll as 'rare reading scrolls', but lease and session are not discussed at all and mode is merely restated as 'mode=on|off|status'.

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 resource and effect: opt-in AMBIENT activity between verbs, emitting low-rate pointer drifts and rare reading scrolls while the session idles. It is clearly distinguishable from a mouse/click tool by the explicit 'never clicks/types/selects/navigates' clause. It does not, however, differentiate itself from the sibling tools humanize_on, humanize_off and humanize_status, which overlap heavily with its own on/off/status modes.

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?

Usage is implied rather than stated: 'Opt-in ... (default off)' plus 'Fills the zero-mouse-events silence behavioural scorers flag' tells an agent the motivation for enabling it. What is missing is any explicit routing — it never says when to prefer this over humanize_on/humanize_off/humanize_status, nor that those siblings exist, even though they appear to manage the same feature set.

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