Skip to main content
Glama

summarize_report

Read the results.sarif an earlier draugr scan wrote and return its findings ranked by priority, deduplicated, with the rule documentation link for each.

Instructions

Read the results.sarif an earlier draugr scan wrote and return its findings ranked by priority, deduplicated, with the rule documentation link for each. Answers 'what should I fix first?' from a scan that already ran, without starting a new one. A severity gate such as --fail-on high judges each finding's rankedSeverity, not its severity, which is the scanner's rating. A finding the project accepted comes back in accepted, with why, until when and the file and line that record the decision; report it as a decision and do not propose fixing it. The git history of a descriptor rule's line records who wrote it, and report.json carries that history in descriptor.exclusions[].gitHistory. It covers the controls that scan ran and nothing else, so treat it as a floor to build on rather than a complete account of a codebase's security. Given a report.json, read the results.sarif in the same directory, since report.json holds the verdict and the fix list and names each finding only by its fingerprint.

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": "summarize_report"
  }
}

Input Schema

NameRequiredDescriptionDefault
pathYespath to the results.sarif draugr scan wrote; report.json beside it holds the verdict and the fix list, and names each finding only by its fingerprint
limitNomaximum findings to return; defaults to 20
minPriorityNoonly return findings at this priority or above: p1, p2, p3 or p4

Output Schema

NameRequiredDescriptionDefault
nextYesthe first thing to do: of the first actions at or above minPriority in fix_list's lists for this report, the one with the highest priority, the earlier list on a tie; the action's title is quoted
noteNo
feedsNoexploitability datasets the scan consulted; absent means no finding was checked against KEV or EPSS
totalYesfindings the report judged, before any priority filter; excludes suppressed
countsYesevery finding total counts, by rankedSeverity, before any priority filter
acceptedNothe suppressed findings with why, until when and where each decision is written, most severe first, capped at limit; report each as a decision, never as work to fix
findingsYes
returnedYes
suppressedNofindings a recorded decision took out of the ranking; not part of total

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv0.180.0
    • addedOutput schema / properties / findings / items / properties / escalation / properties / cve
      Added value: +{
      +  "description": "the CVE the dataset matched, present where the finding's rule id names none",
      +  "type": "string"
      +}
  2. Changed10 schema fields changedv0.166.3
    • addedOutput schema / properties / accepted / items / properties / rankedSeverity
      Added value: +{
      +  "description": "the severity this finding was ranked at, after the control's minimum severity and every signal; the one the report prints. A suppressed finding fails no gate",
      +  "type": "string"
      +}
    • addedOutput schema / properties / accepted / items / properties / severity / description
      Added value: +"the scanner's own rating, before the control's minimum severity or any signal moved it"
    • changedOutput schema / properties / accepted / items / required
      Previous value: -[
      -  "ruleId",
      -  "severity",
      -  "justification"
      -]New value: +[
      +  "ruleId",
      +  "severity",
      +  "rankedSeverity",
      +  "justification"
      +]
    • addedOutput schema / properties / counts / description
      Added value: +"every finding total counts, by rankedSeverity, before any priority filter"
    • changedOutput schema / properties / findings / items / properties / escalation / properties / from / description
      Previous value: -"the severity the scanner reported"New value: +"the severity before this signal raised it: the scanner's rating, held at severityFloor where that is present"
    • changedOutput schema / properties / findings / items / properties / escalation / properties / to / description
      Previous value: -"the severity the finding was ranked at"New value: +"the severity this signal raised it to"
    • addedOutput schema / properties / findings / items / properties / rankedSeverity
      Added value: +{
      +  "description": "the severity this finding was ranked at (critical, high, medium or low), after the control's minimum severity and every signal; the one the report prints and a severity gate such as --fail-on high or failOnNew judges",
      +  "type": "string"
      +}
    • changedOutput schema / properties / findings / items / properties / severity / description
      Previous value: -"critical, high, medium or low"New value: +"the scanner's own rating (critical, high, medium or low), before the control's minimum severity or any signal moved it; a severity gate reads rankedSeverity, not this"
    • addedOutput schema / properties / findings / items / properties / severityFloor
      Added value: +{
      +  "description": "the minimum severity this finding's control holds its findings at, present only where it raised the scanner's rating",
      +  "type": "string"
      +}
    • changedOutput schema / properties / findings / items / required
      Previous value: -[
      -  "severity",
      -  "ruleId",
      -  "message"
      -]New value: +[
      +  "severity",
      +  "rankedSeverity",
      +  "ruleId",
      +  "message"
      +]
  3. Changed3 schema fields changedv0.166.0
    • changedOutput schema / properties / accepted / description
      Previous value: -"the suppressed findings with who decided and why, most severe first, capped at limit; report each as a decision, never as work to fix"New value: +"the suppressed findings with why, until when and where each decision is written, most severe first, capped at limit; report each as a decision, never as work to fix"
    • removedOutput schema / properties / accepted / items / properties / acceptedBy
      Removed value: -{
      -  "description": "a name the descriptor's deprecated acceptedBy states, which nothing verifies; the git history of the exclusion's line in source records who wrote it",
      -  "type": "string"
      -}
    • addedOutput schema / properties / accepted / items / properties / line
      Added value: +{
      +  "description": "for a rule in the descriptor, the line in source it starts at; report.json carries that line's git history in descriptor.exclusions[].gitHistory",
      +  "type": "integer"
      +}
  4. Changed1 schema field changedv0.162.1
    • changedOutput schema / properties / accepted / items / properties / acceptedBy / description
      Previous value: -"who accepted it; empty means the decision is unattributed"New value: +"a name the descriptor's deprecated acceptedBy states, which nothing verifies; the git history of the exclusion's line in source records who wrote it"
  5. Changed1 schema field changedv0.157.0
    • changedOutput schema / properties / next / description
      Previous value: -"what to do with this result"New value: +"the first thing to do: of the first actions at or above minPriority in fix_list's lists for this report, the one with the highest priority, the earlier list on a tie; the action's title is quoted"
  6. First observedv0.1.0

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 substantial work: it discloses deduplication, priority ranking, that a severity gate judges rankedSeverity rather than the scanner's severity, that accepted findings return with rationale/expiry/file/line and must be reported as decisions not fixes, and that coverage is limited to what the scan ran. It omits error behavior (missing file), pagination interaction with limit, and an explicit read-only statement, which keeps it below a 5.

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?

Purpose is front-loaded in the first sentence, but the text runs six sentences and includes a tangential, hard-to-parse aside about git history of a descriptor rule's line and descriptor.exclusions[].gitHistory. That material is relevant to report.json but is not clearly tied to invoking this tool, so some sentences do not fully earn their place.

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?

An output schema exists, so return values need not be spelled out, and the description still explains the non-obvious semantics an agent needs: rankedSeverity versus severity, the accepted block, and the coverage-floor caveat. The main residual gap is operational behavior on bad or missing input, which is minor for a read tool.

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 description coverage is 100%, so the schema already documents path, limit and minPriority; baseline is 3. The description reinforces the path semantics (results.sarif, report.json beside it, fingerprints as identifiers) but adds nothing about limit or minPriority beyond what the schema states.

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: reads results.sarif from an earlier draugr scan and returns findings ranked by priority, deduplicated, each with its rule documentation link. It explicitly distinguishes itself from the scan sibling ('without starting a new one') and frames the question it answers ('what should I fix first?'), so an agent can select it without inspecting the schema.

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 clear when-to-use context: it is for a scan that already ran, not for launching a new one, and it clarifies that given a report.json you should point it at the results.sarif in the same directory. It does not explicitly route to or exclude closer siblings such as diff_reports or fix_list, so the alternative-selection guidance is not exhaustive.

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