Skip to main content
Glama

triage_alerts

Read-onlyIdempotent

Scan locked dependencies for known vulnerabilities and sort alerts into action, review, or filtered with reasons, so you can see which security issues need attention first.

Instructions

Scan a project's locked dependencies for known vulnerabilities and sort every alert into action (fix now), review (worth a look, not urgent) or filtered (deprioritized: dev-only, test-only...), with the reasons. Use it first, to see which alerts matter; then call plan_fixes for the change that closes them. Works on Python (uv, Poetry, pip requirements) and JavaScript (npm, pnpm) projects. Read-only: runs pip-audit or the package manager's audit, which query their advisory databases over the network, plus EPSS and CISA KEV lookups unless offline=true.

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>"
    },
    "name": "triage_alerts"
  }
}

Input Schema

NameRequiredDescriptionDefault
pathYesAbsolute path of the project folder: the one holding pyproject.toml, uv.lock, poetry.lock, requirements*.txt or package.json.
offlineNoSkip the EPSS and CISA KEV downloads (cached copies are still used). The scanner itself still needs its advisory database.

Output Schema

NameRequiredDescriptionDefault
okYesfalse when the scan could not run; see error.
nextNoThe suggested next call, when there is one.
errorNo
notesNoScanner warnings, if any.
totalNoNumber of alerts.
alertsNo
countsNoAlerts per verdict.
detailsNoThe end of depquiet's stderr.
projectNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed7 schema fields changedv0.1.1
    • addedInput schema / properties / offline / description
      Added value: +"Skip the EPSS and CISA KEV downloads (cached copies are still used). The scanner itself still needs its advisory database."
    • addedInput schema / properties / path / description
      Added value: +"Absolute path of the project folder: the one holding pyproject.toml, uv.lock, poetry.lock, requirements*.txt or package.json."
    • addedOutput schema / $defs
      Added value: +{
      +  "Alert": {
      +    "description": "One vulnerability alert. Empty fields are left out.",
      +    "properties": {
      +      "cve": {
      +        "description": "Advisory ID: a CVE, GHSA or PYSEC ID.",
      +        "title": "Cve",
      +        "type": "string"
      +      },
      +      "dependency_type": {
      +        "description": "direct, transitive or unknown.",
      +        "title": "Dependency Type",
      +        "type": "string"
      +      },
      +      "epss_score": {
      +        "description": "EPSS exploit probability, 0 to 1.",
      +        "title": "Epss Score",
      +        "type": "number"
      +      },
      +      "fix_package": {
      +        "description": "The direct dependency to upgrade, when it differs from package.",
      +        "title": "Fix Package",
      +        "type": "string"
      +      },
      +      "fix_version": {
      +        "description": "First version with the fix.",
      +        "title": "Fix Version",
      +        "type": "string"
      +      },
      +      "introduced_by": {
      +        "description": "Dependency path to the package.",
      +        "items": {
      +          "type": "string"
      +        },
      +        "title": "Introduced By",
      +        "type": "array"
      +      },
      +      "is_dev": {
      +        "title": "Is Dev",
      +        "type": "boolean"
      +      },
      +      "package": {
      +        "description": "The vulnerable package.",
      +        "title": "Package",
      +        "type": "string"
      +      },
      +      "python_scope": {
      +        "description": "Like \"<3.11.5\": only Pythons below it install the vulnerable copy.",
      +        "title": "Python Scope",
      +        "type": "string"
      +      },
      +      "reasons": {
      +        "description": "Why the alert got that verdict.",
      +        "items": {
      +          "type": "string"
      +        },
      +        "title": "Reasons",
      +        "type": "array"
      +      },
      +      "severity": {
      +        "title": "Severity",
      +        "type": "string"
      +      },
      +      "title": {
      +        "title": "Title",
      +        "type": "string"
      +      },
      +      "url": {
      +        "title": "Url",
      +        "type": "string"
      +      },
      +      "verdict": {
      +        "description": "action, review or filtered.",
      +        "title": "Verdict",
      +        "type": "string"
      +      },
      +      "version": {
      +        "description": "The locked version that is vulnerable.",
      +        "title": "Version",
      +        "type": "string"
      +      }
      +    },
      +    "title": "Alert",
      +    "type": "object"
      +  }
      +}
    • removedOutput schema / additionalProperties
      Removed value: -true
    • addedOutput schema / properties
      Added value: +{
      +  "alerts": {
      +    "items": {
      +      "$ref": "#/$defs/Alert"
      +    },
      +    "title": "Alerts",
      +    "type": "array"
      +  },
      +  "counts": {
      +    "additionalProperties": {
      +      "type": "integer"
      +    },
      +    "description": "Alerts per verdict.",
      +    "title": "Counts",
      +    "type": "object"
      +  },
      +  "details": {
      +    "description": "The end of depquiet's stderr.",
      +    "title": "Details",
      +    "type": "string"
      +  },
      +  "error": {
      +    "title": "Error",
      +    "type": "string"
      +  },
      +  "next": {
      +    "anyOf": [
      +      {
      +        "type": "string"
      +      },
      +      {
      +        "type": "null"
      +      }
      +    ],
      +    "description": "The suggested next call, when there is one.",
      +    "title": "Next"
      +  },
      +  "notes": {
      +    "anyOf": [
      +      {
      +        "type": "string"
      +      },
      +      {
      +        "type": "null"
      +      }
      +    ],
      +    "description": "Scanner warnings, if any.",
      +    "title": "Notes"
      +  },
      +  "ok": {
      +    "description": "false when the scan could not run; see error.",
      +    "title": "Ok",
      +    "type": "boolean"
      +  },
      +  "project": {
      +    "title": "Project",
      +    "type": "string"
      +  },
      +  "total": {
      +    "description": "Number of alerts.",
      +    "title": "Total",
      +    "type": "integer"
      +  }
      +}
    • addedOutput schema / required
      Added value: +[
      +  "ok"
      +]
    • changedOutput schema / title
      Previous value: -"triage_alertsDictOutput"New value: +"TriageResult"
  2. First observedv0.1.0

TDQS

A4.4/5.0
Behavior4/5

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

Annotations already declare readOnlyHint, openWorldHint and idempotentHint, but the description adds real specifics: it runs pip-audit or the package manager's audit, queries advisory databases over the network plus EPSS and CISA KEV, and explains what offline=true actually skips versus what still needs the network. This matters because a user could wrongly assume offline makes the tool fully local.

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?

Front-loaded with the core action and the sequencing advice, with ecosystem and network caveats trailing as they should. The final sentence is dense with parentheticals and stacked clauses, which costs a little readability but no sentence is filler.

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 two-parameter read-only scanner with rich annotations and an output schema, the description covers everything an agent needs: what it does, the triage buckets, the ordering relative to plan_fixes, supported ecosystems, and the network/offline behavior. Return-format detail is correctly delegated to the output schema.

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% and the description adds little beyond it, restating the offline exception and listing the manifest/lockfile types that path can point to. That ecosystem list is mildly useful corroboration of what path accepts, but the schema already carries the semantics, so the 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 concrete verb and resource ('scan a project's locked dependencies for known vulnerabilities') plus the output shape (alerts sorted into action/review/filtered with reasons). It also explicitly separates itself from the sibling by saying plan_fixes is the follow-up step, so an agent can tell the two apart without opening either schema.

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

Usage Guidelines5/5

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

Gives an explicit ordering rule: 'Use it first, to see which alerts matter; then call plan_fixes for the change that closes them.' Both the when-to-use condition and the alternative sibling are named, leaving nothing to inference.

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