Skip to main content
Glama

prompt_injection_checklist

Read-onlyIdempotent

Map prompt injection paths and controls for an agent from its tools and untrusted sources, then produce a status checklist to fill—not a scan.

Instructions

Enumerate injection paths and controls for an agent from its tools and untrusted data sources; produces a checklist with status to fill, not a scan.

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": {
      "agent": "<agent>",
      "untrustedSources": "<untrustedSources>",
      "tools": "<tools>",
      "humanInLoop": "<humanInLoop>"
    },
    "name": "prompt_injection_checklist"
  }
}

Input Schema

NameRequiredDescriptionDefault
agentYes
toolsYes
humanInLoopYes
untrustedSourcesYes

Output Schema

NameRequiredDescriptionDefault
kindNo
agentNo
pathsNo
sourcesNo
blockersNo
controlsNo
validatedNo
limitationNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv0.5.0
    • removedInput schema / properties / tools / items / additionalProperties
      Removed value: -false
    • changedInput schema / properties / tools / items / required
      Previous value: -[
      -  "name",
      -  "mutating",
      -  "external"
      -]New value: +[
      +  "name",
      +  "mutating"
      +]
  2. First observedv0.4.1

TDQS

B3.4/5.0
Behavior4/5

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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, so the safety profile is covered. The description adds genuinely new behavioral context beyond those annotations: the result is a checklist with statuses left for the caller to fill in, not an automated scan, which sets expectations about completeness of the output. It does not mention determinism or limits (e.g. max 40 tools), but that is largely captured by the schema.

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?

A single compact sentence with the core purpose front-loaded and the clarifying caveat appended. No filler or repetition, though the semicolon clause about 'checklist vs scan' could be tightened without losing meaning.

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?

An output schema exists, so return values need not be explained, and annotations cover the safety profile. What is missing for a four-required-parameter tool with a nested tools array is guidance on what to supply for humanInLoop and how tool attributes (external/mutating) affect the checklist. Adequate but with clear gaps for correct invocation.

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

Parameters2/5

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

Schema description coverage is 0%, so the description must carry parameter meaning. It only loosely gestures at three inputs ('agent', 'tools', 'untrusted data sources') and says nothing about humanInLoop, which is required, or about the nested per-tool fields 'external' and 'mutating' that materially change the injection analysis. The enum of untrusted source types is only discoverable from the schema.

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: 'Enumerate injection paths and controls for an agent from its tools and untrusted data sources.' The clause 'produces a checklist with status to fill, not a scan' further pins down what the output is, distinguishing it from scanner-style siblings such as threat_model or mcp_server_review. It stops short of naming any sibling explicitly, so a 5 is not warranted.

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

Usage Guidelines3/5

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

The phrase 'not a scan' implicitly tells the agent this is a planning/enumeration step rather than an automated vulnerability scan, which is useful routing signal. However, there is no explicit when-to-use condition, no named alternative among the many adjacent siblings (threat_model, mcp_server_review, agent_tool_contract), and no statement of prerequisites or when to skip it.

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