Skip to main content
Glama
@caelaralabs/mcp

@caelaralabs/mcp

caelara_agent_agent_input_firewall

Scans untrusted input for injection risk, redacts PII, and returns sanitized text with an allow, review, or block verdict.

Instructions

[Agent Safety] Scan untrusted input: injection risk + PII redaction + sanitized text + allow/review/block verdict Measured p50 11 ms, p95 15 ms (2026-11-21, benchmarks.caelaralabs.com). Calls https://agent.caelaralabs.com.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "text": "<text>"
    },
    "name": "caelara_agent_agent_input_firewall"
  }
}

Input Schema

NameRequiredDescriptionDefault
textYesUntrusted input text
redact_piiNoRedact PII in sanitized output (default true)

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.0.1

TDQS

B3.2/5.0
Behavior3/5

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

With no annotations, the description carries the burden and does add useful operational context: it discloses the verdict taxonomy (allow/review/block) and measured latency (p50 11 ms, p95 15 ms with a dated source), which lets an agent budget calls. However it never states that the operation is side-effect free, whether submitted text is retained or logged (critical for a tool handling PII), or what the endpoint call implies about credentials.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two compact sentences with the purpose front-loaded and the latency/endpoint details trailing where they belong. The benchmark parenthetical with a date and URL is slightly heavy for the value it carries, but it is short and earns its place as a performance signal.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

There is no output schema, so the description is the only source for return values; it names the outputs but gives no shape for them (e.g., whether injection risk is a score, label, or list of findings). For a composite tool it is adequate to select and call, but not to interpret the response.

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

Parameters3/5

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

Schema coverage is 100%, so both parameters are already documented, and the description only echoes that PII redaction feeds the sanitized output. It adds no syntax, size limits, or guidance on when to set redact_pii=false, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Scan untrusted input') and enumerates the four things produced: injection risk, PII redaction, sanitized text, and an allow/review/block verdict. It is clearly a composite safety gate, but it never distinguishes itself from the single-purpose siblings such as caelara_agent_prompt_injection_scanner or caelara_agent_pii_redactor, so an agent cannot tell from the text alone why it should pick this one.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The phrase 'untrusted input' hints at the trigger condition, but there is no explicit when-to-use statement, no when-not-to-use, and no routing to the overlapping prompt-injection and PII sibling tools. For a composite tool in a catalog full of narrower scanners, the absence of that routing is a real gap.

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

Deploy Server

Other Tools