Runs an eight-stage isolation test against your staging Supabase and issues a signed attestation.
Scored across 6 tools
Each tool has a distinct role: list_projects, run_review, get_run_status, wait_for_run, get_findings, and get_report. The main overlap is between get_run_status and wait_for_run, which return the same status information, but the descriptions clearly differentiate them by blocking behavior and the get_findings/get_report split is explicitly clarified.
All six tools follow a consistent verb_noun snake_case pattern (get_findings, get_report, get_run_status, list_projects, run_review, wait_for_run). No mixing of conventions or vague verbs.
Six tools is well-scoped for the domain of running and retrieving authorization test results on Supabase projects. Each tool earns its place in the discover-run-poll-retrieve workflow with no redundancy.
The lifecycle is well covered: list projects, start a run, poll status, wait, and retrieve both compact findings and the full signed report. Minor gaps exist, such as no cancel/abort tool and no explicit list of past runs, but core workflows are complete.
Returns the cells that did not pass in a complete run that has a signed report — verdict, surface, operation, boundary, identity, layer outcomes and notes — plus tenancy flags, blocking findings and, when the report includes catalog hygiene checks, the hygiene block (the checks run, each finding's check, severity, object, detail and hint, and any check not run). For a complete run with no signed report, while one can still be bought on the run page, it returns the run's unverified result instead: the verdict, surface, operation and principal of each cell that did not pass, the run page's finding lines, the counts, the coverage lines, the access model summary and the same hygiene block. The list of cells is empty when every tested cell passed. Cell results contain no fix suggestions: the engine records only what it observed. Hygiene findings are catalog observations, never verdicts; each carries the engine's one-line hint for the project owner's review.
| Name | Required | Description | Default |
|---|---|---|---|
| run_id | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already establish readOnlyHint=true and openWorldHint=false, yet the description still adds genuine behavioral context: the cell list is empty when every tested cell passed, cell results contain no fix suggestions, and hygiene findings are catalog observations rather than verdicts. The conditional switch between signed-report and unverified output is a meaningful disclosure the annotations cannot express. It stops short of covering auth needs, errors, or pagination.
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 content is delivered as a dense, run-on paragraph with long chains of enumerated fields joined by dashes and commas, forcing the agent to parse a wall of text to find the branch logic. The main case is front-loaded, but the structure does not separate the two output modes, and much of the field enumeration is padding that could be tightened.
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 carries the full return-value burden and largely meets it: it describes both output shapes, the edge case of an empty cell list, and the nature of hygiene findings. Given a single required parameter and a simple schema, the main remaining gaps are error behavior and run_id sourcing, which are minor.
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 one parameter (run_id) with 0% schema description coverage, so the description must carry the semantics. It contextualizes run_id indirectly by tying it to a complete run with or without a signed report, but never defines the identifier's format, source, or valid values. Self-evident enough to be workable, but it leaves the gap the schema created unfilled.
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 (the cells that did not pass in a run) and enumerates the payload blocks returned (verdict, surface, operation, boundary, identity, layer outcomes, tenancy flags, blocking findings, hygiene block), so the agent knows what it gets. It never differentiates itself from siblings like get_report or get_run_status, and the two-branch return shape (signed report vs. unverified result) makes the core purpose slightly harder to grasp than it needs to be.
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 only implied: the description explains the two run states that change the output (complete run with a signed report vs. a complete run where one can still be bought), but never states when an agent should call get_findings rather than get_report or get_run_status. No exclusions or prerequisites are given.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Returns the full signed report of a complete run as the exact stored JSON, with its verify_url and report_sha256 (sha256 of the canonical signed bytes, the fingerprint shown on the verify page). Use get_findings for the compact view.
| Name | Required | Description | Default |
|---|---|---|---|
| run_id | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds real value beyond that: it discloses the return shape (exact stored JSON plus verify_url and report_sha256) and defines what report_sha256 actually hashes, which is verifiable behavioral context. It does not say what happens if the run is not yet complete.
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 sentences, zero waste, front-loaded with what is returned before the alternative-tool routing. The parenthetical defining report_sha256 earns its space by making the returned field interpretable.
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 usefully documents the key return fields, and annotations cover the read-only nature. The remaining gap is behavioral edge cases (incomplete run, unknown run_id) rather than core invocation information.
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 run_id parameter is undocumented in both schema and description. The phrase 'of a complete run' implicitly ties run_id to a run identifier, but no format, source, or example is given, so the description only marginally compensates 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?
States a specific verb and resource ('Returns the full signed report of a complete run') and explicitly distinguishes itself from the sibling get_findings by contrasting full vs. compact views. An agent can tell it apart from get_findings without opening either 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?
Names the alternative explicitly ('Use get_findings for the compact view'), giving a clear condition for choosing between the two. It stops short of stating exclusions such as what happens when the run is not complete, so it is clear but not fully exhaustive.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Returns a run's status (queued, running, complete or failed), its current stage, timestamps and, when a signed report exists, its verify_url.
| Name | Required | Description | Default |
|---|---|---|---|
| run_id | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description does add real behavior beyond that: verify_url is only present when a signed report exists, a genuine conditional disclosure. It omits the key behavioral trait that separates it from wait_for_run — whether this call returns immediately or waits.
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 dense sentence, front-loaded with the primary return value and the enumerated statuses, then the secondary fields, then the conditional. No filler or redundancy.
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 describe return values and it does so reasonably (status enum, stage, timestamps, conditional verify_url). However, it lacks polling/freshness guidance and any disambiguation from wait_for_run, which is what an agent most needs in this sibling set.
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 run_id parameter is documented nowhere. 'A run's status' implies the parameter identifies a run, but format, source, or acceptance of partial identifiers is never stated, so the description 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?
States a specific verb (returns) and resource (a run's status) and even enumerates the status values and the other returned fields (stage, timestamps, verify_url). That is far more than a restatement of the name. It does not, however, differentiate itself from the sibling wait_for_run, which an agent would plausibly confuse it with.
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 or when-not-to-use guidance at all. The sibling set contains wait_for_run and get_report, and the description never says whether this is a non-blocking snapshot versus a blocking wait, or when to prefer get_report over this.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Lists the Supabase projects connected to this account with id, name, connection state and whether each is ready to run. Call this first to get a project_id.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds the account-scoping constraint and the entry-point workflow context, but does not disclose pagination, rate limits, or other behavioral details beyond the annotations.
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 tightly written sentences with zero waste; the purpose and output fields come first, followed by the workflow instruction. It is appropriately sized for a simple list 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?
Given zero parameters, rich annotations, and no output schema, the description provides everything needed: what is listed, what fields are returned, and that it should be called first to obtain a project_id.
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 has zero parameters, so the baseline is 4. The description adds no parameter information because none exists.
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 (Lists) and resource (Supabase projects connected to this account), and enumerates the returned fields (id, name, connection state, readiness). This clearly distinguishes it from the sibling tools, which operate on findings, reports, and runs rather than projects.
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?
Explicitly instructs 'Call this first to get a project_id,' giving clear workflow ordering. It does not include when-not conditions or name alternatives, but no sibling tool provides a competing project-listing capability.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Starts an authorization test run on a connected project and returns the run_id and status. If the project already has a run waiting to start, returns that run instead of starting another. Runs that do not fail count against the monthly allowance.
| Name | Required | Description | Default |
|---|---|---|---|
| project_id | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations declare a non-read-only, non-idempotent, non-destructive write; the description adds two behavioral facts annotations cannot convey: the queue-dedupe behavior (an existing waiting run is returned instead of a new one) and the monthly-allowance consumption rule for non-failing runs. No auth or failure-handling detail, but this is meaningfully beyond the structured hints.
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, front-loaded with the action and return values, then the dedupe edge case, then the quota caveat. Every sentence contributes; the quota sentence is slightly tangential but still operational knowledge an agent needs.
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 action tool with no output schema, the description compensates by naming the return fields, the dedupe behavior, and the quota impact. The main remaining gap is what happens if a run fails or how the project must be connected, but nothing critical to invoking it correctly 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?
One required parameter with 0% schema description coverage, so the description must carry the burden. 'On a connected project' hints that project_id must reference a connected project, but the description adds no format, source, or lookup guidance (e.g., list_projects). Only marginal value over the bare 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?
States a specific verb and resource ('Starts an authorization test run on a connected project') and names the concrete return values (run_id and status). An agent can immediately separate this from get_run_status, wait_for_run, and list_projects without opening any 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 dedupe rule ('if the project already has a run waiting to start, returns that run instead') gives useful implied usage context, but the description never states when to prefer this tool over siblings like get_run_status or wait_for_run, nor any prerequisites such as needing a connected project. Usage is implied rather than guided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Waits up to timeout_seconds (1 to 60) for a run to finish, then returns the same information as get_run_status. Call it again while the status is still queued or running.
| Name | Required | Description | Default |
|---|---|---|---|
| run_id | Yes | ||
| timeout_seconds | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations only declare readOnlyHint=true and openWorldHint=false; the description adds the blocking/latency profile (waits up to a bounded timeout) and the required re-invocation loop, which the annotations cannot convey. It does not discuss failure modes such as what is returned on timeout expiry, which is the remaining behavioral gap.
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 sentences, no filler, with the blocking behavior and timeout bound front-loaded ahead of the retry instruction. Every clause 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?
For a simple two-parameter read tool, the definition covers purpose, the retry contract, and the return payload by reference to get_run_status, which the agent can resolve from the sibling set. Polling interval and timeout-expiry return behavior are the only omissions.
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% and the schema gives only bare types, so the description must carry the load: it supplies the timeout_seconds domain (1 to 60), which is genuinely new information not in the schema. run_id is left implicit, but its meaning ('the run to wait on') is self-evident from the resource name.
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 ('waits ... for a run to finish') and immediately distinguishes itself from the sibling get_run_status by noting it returns the same payload but blocks. An agent can tell the two apart without opening either 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?
'Call it again while the status is still queued or running' gives the operating loop and implicitly contrasts with one-shot status polling via get_run_status. It stops short of an explicit 'use this instead of repeatedly calling get_run_status' rule, so it is clear context without stated exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_findingsget_reportget_run_statuslist_projectsrun_reviewwait_for_runScan a deployed app URL for exposed keys, open Supabase tables and missing security headers.
Signed x402 attestations and bounty-packet preflight checks, paid per call in USDC on Base.
Attestation infrastructure for the agentic economy: signed, independently verifiable verdicts.
Probe a signup URL you own and score whether an AI agent can sign up unaided.
Add one secure layer between your agents and this server.