Agent Tools

dependency_audit

Audit a whole project before an agent installs or runs it: JSON body {manifest, type, readme}. manifest is package.json (npm) or requirements.txt (PyPI) text. Checks up to 25 dependencies against OSV.dev advisories, deprecation and latest versions, and screens the optional README for prompt injection. Returns one safe, caution or avoid verdict plus a ranked fix list (upgrade to version X). Price $0.05 USDC per call via x402. Example output: {"verdict":"avoid","ecosystem":"npm","manifest_type":"package.json","summary":{"dependencies_found":2,"checked":2,"not_checked_over_limit":0,"lookup_failed":0,"avoid":1,"caution":0,"known_vulnerabilities":5},"fixes":[{"priority":"high","package":"lodash","action":"upgrade to >= 4.17.21","why":["5 kn

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": {
      "manifest": "<manifest>"
    },
    "name": "dependency_audit"
  }
}

Input Schema

Name	Required	Description
`type`	No	Optional. Auto-detected if omitted
`readme`	No	Optional README or docs text to screen for prompt injection, max 50,000 chars
`manifest`	Yes	package.json or requirements.txt contents as a string, max 200,000 chars

Schema Changelog

Changes observed during successful MCP inspections.

AddedOct 30, 2026

TDQS

A4/5.0

Behavior4/5

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

With no annotations, the description carries the full burden and does it well: it discloses the 25-dependency cap, the OSV.dev data source, the 200k/50k input limits via schema, the three-way verdict, and — critically — the $0.05 USDC x402 payment requirement, which is a cost gate an agent must know before calling. It stops short of explaining behavior on invalid/oversized manifests or lookup failures beyond the example's 'lookup_failed' counter.

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

Conciseness3/5

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

The operational facts are front-loaded, which is good, but the description then appends a large example-output JSON that is truncated mid-string ('"why":["5 kn'), consuming tokens without delivering complete information. The example could be dropped or shortened to one line without losing meaning.

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

Completeness4/5

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

There is no output schema, but the description compensates by showing the return shape (verdict, ecosystem, summary counters, ranked fixes) and the pricing model, which is what an agent needs to decide and to parse results. The precise semantics of 'safe' vs 'caution' vs 'avoid' thresholds remain unstated.

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 all three parameters are already documented in structured form, and the description mostly restates them ('manifest is package.json (npm) or requirements.txt (PyPI) text'). It adds no detail on the enum auto-detection logic or on how the README is screened, so baseline 3 applies.

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 verb and resource with explicit scope: 'Audit a whole project before an agent installs or runs it', covering up to 25 dependencies against OSV.dev plus deprecation/latest-version checks and README injection screening. The 'whole project' framing plus the composite checks distinguishes it from siblings like package_advisories (single package) and prompt_injection_check (README-only), even though neither is named.

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?

Gives a clear use context ('before an agent installs or runs it'), which is exactly the decision point an agent needs. It does not name alternatives or state when-not to use it (e.g., for a single package, use package_advisories), so it falls short of a 5.

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

All Agent Tools tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?