Skip to main content
Glama
Sign In
kr.ai.vdb/vdb

kr.ai.vdb/vdb

by 0pstech
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

vdb_harden

Analyze Python code to determine if user input reaches dangerous operations through transitive dependencies, without relying on CVEs. Get actionable paths, a non-invasive call-site fix, and residual risk.

Instructions

Decide whether attacker-controlled data can reach a dangerous operation through this file's transitive dependencies — with no CVE required. Use it on code that passes user input into a third-party API. The file is abstracted LOCALLY first (identifiers renamed, literals reduced to shapes, bodies dropped); only that abstraction and the lockfile are sent, never source text. Returns decided paths, a call-site fix that does not modify the dependency, and the residual risk the fix does not cover.

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": {
      "path": "<path>",
      "manifest_path": "<manifest_path>"
    },
    "name": "vdb_harden"
  }
}

Input Schema

NameRequiredDescriptionDefault
pathYesPython file to analyze.
manifest_pathYesuv.lock / poetry.lock / Pipfile.lock / requirements.txt / CycloneDX. Version ranges cannot be analyzed — the answer differs per resolved version.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.2.0

TDQS

A4.8/5.0
Behavior5/5

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

Discloses important behavioral details beyond the schema: the file is abstracted locally, only the abstraction and lockfile are transmitted, source text is never sent, and the result includes a non-modifying call-site fix plus residual risk. Since annotations are absent, the description carries the full transparency burden and satisfies it.

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

Conciseness5/5

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

Three focused sentences: what it decides, when to use it, and how it behaves safely. Every sentence contributes distinct information with no filler, and the main purpose is front-loaded.

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

Completeness5/5

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

For a tool with no output schema, it still summarizes the return value (decided paths, call-site fix, residual risk) and covers privacy, input constraints, and dependency resolution limitations. An agent has enough context to select and invoke the tool correctly.

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

Parameters5/5

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

Adds meaningful semantics beyond the schema, especially for manifest_path: it names the accepted lockfile formats and warns that version ranges cannot be analyzed because the answer depends on the resolved version. This prevents a common misuse that the schema alone would not catch.

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

Purpose5/5

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

States a specific analytic goal: deciding whether attacker-controlled data can reach a dangerous operation through transitive dependencies. The 'no CVE required' qualifier sharply distinguishes it from CVE-oriented sibling tools like vdb_lookup and vdb_scan_lockfile.

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

Usage Guidelines4/5

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

Explicitly directs use to 'code that passes user input into a third-party API,' which tells an agent when to apply it. It does not name sibling alternatives by exclusion, but the use context is clear enough to guide selection.

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