Propose selector repairs by matching exact accessible identities in a live snapshot; it never executes or modifies, and ambiguous or fuzzy identities return no automatic repair.
Propose selector repairs using exact accessible identities in a real snapshot. Does not execute or modify anything; ambiguous/fuzzy identity yields no automatic repair.
| Name | Required | Description | Default |
|---|---|---|---|
| locator | Yes | ||
| sessionId | Yes |
Changes observed during successful MCP inspections.
v0.4.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and destructiveHint=false, so 'does not execute or modify anything' is largely a restatement. The one genuine addition is the failure mode: ambiguous or fuzzy identities yield no automatic repair, which tells the agent to expect empty/absent results rather than guessing. Return shape and how 'real snapshot' is resolved remain undisclosed.
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, front-loaded with the purpose followed by the safety/limitation caveat. Nothing is padded, though the second sentence spends half its length repeating annotation-declared read-only behavior.
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 exists, so the description must carry the burden of explaining what a suggested repair looks like and how it relates to sessionId and the eleven-field locator. It omits the prerequisite that a snapshot must exist and says nothing about return values, leaving an agent unable to predict output or parameter trade-offs.
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 nested locator object exposes eleven mutually alternative fields (css, role, name, testId, automationId, etc.) with no guidance anywhere. The description never mentions sessionId, the locator object, or which identity fields take precedence, so it compensates for none of the documentation 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?
States a specific verb ('Propose') and resource ('selector repairs') plus the mechanism ('exact accessible identities in a real snapshot'). It implicitly contrasts with heal_apply's apply semantics but never names that sibling, so an agent must infer the distinction.
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 'Does not execute or modify anything' clause hints this is the dry-run counterpart to a mutation tool, and the ambiguity caveat hints when it will fail, but neither states when to choose this over heal_apply nor what prerequisite state (e.g. an existing snapshot/session) is needed.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.