Agentifact - How We Score

Scoring Methodology

Every listing in Agentifact is scored across five dimensions on a 0-100 scale. Scores are derived from direct testing, not vendor self-reporting. The composite trust score is the unweighted average of all five dimensions.

What "Verified" means

A "Verified" badge means Agentifact has directly tested the tool's API endpoint or integration, checked the documentation against actual behavior, and confirmed the scoring dimensions are current. Verified status is not self-reported by vendors.

Stale policy

Listings are automatically flagged Stale when more than 90 days have passed since the last verified date. We re-verify stale listings on a rolling schedule, highest traffic first.

Score color guide

90-100
Highly trusted
70-89
Trusted
40-69
Caution
0-39
Not recommended

The five dimensions

Agent Readiness20%

Can an AI agent reliably call this tool without supervision? Measures API quality, structured output schemas, error codes that are machine-parsable, retry behavior, and whether the tool exposes an OpenAPI or MCP-compatible interface.

ScoreWhat it means
0-19No machine interface. Requires human configuration.
20-39Basic REST API. Inconsistent error codes. No schema.
40-59Documented API with some structured outputs. Partial schema.
60-79OpenAPI spec available. Predictable errors. Retryable endpoints.
80-89MCP or A2A compatible. Full schema. Rate limit headers exposed.
90-100Native MCP server or A2A endpoint. Structured outputs, retry signals, tool manifest.
Trust20%

Is this tool trustworthy as an infrastructure component? Covers ownership verification, documentation currency, known incident history, changelog maintenance, and whether the vendor actively maintains the integration.

ScoreWhat it means
0-19Anonymous vendor. No changelogs. Known security incidents unaddressed.
20-39Verified vendor identity. Sparse docs. No public incident record.
40-59Active maintenance. Some documentation gaps. Minor incidents disclosed.
60-79Regular releases. Clear ownership. Docs updated within 90 days.
80-89SOC 2 or equivalent. SLA published. Incident post-mortems public.
90-100Verified ownership, active maintenance, public SLA, disclosed incidents, enterprise trust signals.
Interoperability20%

How many protocols and integration surfaces does this tool expose? Scores MCP support, A2A compatibility, REST API coverage, webhook support, and OpenAPI specification completeness.

ScoreWhat it means
0-19Single proprietary integration only.
20-39Basic REST. No standard protocols.
40-59REST + webhook. Partial protocol coverage.
60-79MCP or A2A support. REST documented with OpenAPI.
80-89MCP + REST + webhooks. A2A compatible. SDK available.
90-100Full MCP + A2A + REST. OpenAPI spec. Multiple SDKs. Embeddable tool manifest.
Security20%

What security controls protect this tool when called by an agent? Evaluates authentication mechanisms, PII handling policies, data retention practices, audit logging, and compliance certifications.

ScoreWhat it means
0-19No auth. Unknown PII policy. No audit trail.
20-39API key auth only. Basic privacy policy. No logs.
40-59OAuth or JWT. Privacy policy published. Basic logging.
60-79mTLS or PKCE. Explicit PII controls. Structured audit logs.
80-89Enterprise auth (SSO, RBAC). PII masking configurable. Logs exportable.
90-100SOC 2 Type II, HIPAA, or GDPR certified. Fine-grained RBAC. Immutable audit trail. Data residency controls.
Documentation20%

Can an AI agent use this tool based on its documentation alone without human interpretation? Scores machine-parsable docs, runnable examples, error message quality, and changelog completeness.

ScoreWhat it means
0-19No docs. Or docs exist but are unmaintained.
20-39README only. No runnable examples.
40-59Docs site. Some examples. Changelog sparse.
60-79Full API reference. Code examples in 2+ languages. Active changelog.
80-89OpenAPI spec + interactive playground. Machine-parsable error catalog.
90-100OpenAPI + AsyncAPI spec. Runnable examples. Error catalog. Changelog with breaking change notices. SDK typedocs.

Approval Control Mode

This field is separate from trust score. It answers: "If I give this to my agent, will it run fully autonomously, or require human confirmation?"

FULL AUTO

Can be delegated to an AI agent with no human confirmation.

NEEDS APPROVAL

Requires at least one approval gate before execution.

HUMAN IN LOOP

Human involvement is required through the workflow.

N/A

Approval mode is not applicable for this tool type.

Machine-queryable endpoint

GET /api/tools

Query parameters:

categoryMCP_SERVER | HITL_PROVIDER | A2A_AGENT | FRAMEWORK
approvalModeFULL_AUTO | REQUIRES_APPROVAL | HUMAN_IN_LOOP
minScore0-100 - minimum composite trust score
mcptrue - only MCP-compatible tools
workflowbooking | data-processing | customer-service | ...
limit1-100 (default 20)
← Back to AgentifactTry the API →