Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

io.github.c64dos-png/vector-mirror

NPM · VECTOR-MIRROR · SCANNED SEP 20

Deterministic SVG perception for LLM agents — measures, never guesses. MCP server.

Available components

+1 this week 80 Trust /100
Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →

Supply Chain Security98
  • No malware found by supply-chain analysis.Pass
  • No known CVEs affecting this package version or its production dependencies.Pass
  • No install/post-install scripts declared.Pass
  • 41 of 147 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency45
Schema Quality & AI Usability61
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 2834 tokens (~314/item across 9 items; 9 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management87
  • Stability observed for 26 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage100
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 100% of tool parameters carry a description.Pass
  • Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • We read all 9 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
  • An AI judge read all 10 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Install

How do I install the io.github.c64dos-png/vector-mirror MCP server?

io.github.c64dos-png/vector-mirror runs locally as an npm package, launched with npx -y vector-mirror. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

npm · vector-mirror

# add to Claude Code
claude mcp add c64dos-png-vector-mirror -- npx -y vector-mirror
// .cursor/mcp.json
{
  "mcpServers": {
    "c64dos-png-vector-mirror": {
      "command": "npx",
      "args": [
        "-y",
        "vector-mirror"
      ]
    }
  }
}
// .vscode/mcp.json
{
  "servers": {
    "c64dos-png-vector-mirror": {
      "command": "npx",
      "args": [
        "-y",
        "vector-mirror"
      ]
    }
  }
}
# add to Codex CLI
codex mcp add c64dos-png-vector-mirror -- npx -y vector-mirror
// opencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "c64dos-png-vector-mirror": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "vector-mirror"
      ],
      "enabled": true
    }
  }
}
# add to OpenClaw
openclaw mcp add c64dos-png-vector-mirror --command npx --arg -y --arg vector-mirror
# ~/.hermes/config.yaml
mcp_servers:
  c64dos-png-vector-mirror:
    command: "npx"
    args: ["-y", "vector-mirror"]
// ~/.netclaw/config/netclaw.json
{
  "McpServers": {
    "c64dos-png-vector-mirror": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "vector-mirror"
      ]
    }
  }
}
# add to Vellum
assistant mcp add c64dos-png-vector-mirror -t stdio -c npx -a -y vector-mirror
// mcp.json
{
  "mcpServers": {
    "c64dos-png-vector-mirror": {
      "command": "npx",
      "args": [
        "-y",
        "vector-mirror"
      ]
    }
  }
}
Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 20 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

  • 19 Sept 26 −2
    • Stability: pass → 0.83 functional
  • 18 Sept 26 0
    • Stability: 0.97 → pass security
  • 17 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 93 to 97. That category is still filling its 30-day observation window: 28 days of observed history at the previous scan, 29 at this one. The score rises as the window fills, whether or not the server changes.

  • 14 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

  • 12 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 77 to 80. That category is still filling its 30-day observation window: 23 days of observed history at the previous scan, 24 at this one. The score rises as the window fills, whether or not the server changes.

  • 11 Sept 26 −3
    • Stability: pass → 0.77 functional
  • 10 Sept 26 0
    • Stability: 0.97 → pass security
Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 20 Sept 2026 · Analysed npm/vector-mirror@1.0.0

Provenance No attestation

The registry publishes no build provenance for this version, so there is nothing to verify.

Result No attestation
Ecosystem npm

Background: How many MCP packages publish verified provenance →

Dependencies 147 packages
Packages resolved 147
Stale 41
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 9 exposed · ~2,254 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
vector_mirror_analyze ~424

Analyzes an SVG against spatial constraints, returns a Spotter-Report. Input: SVG + Constraints "#subject TYPE #reference [value]", z.B. "#logo CENTERED-IN #frame"; Typen: vector_mirror_constraints; previousIssueCount → Konvergenz. Output: status PASS/FAIL/PARTIAL, corrections (fix), unchecked, diff; analysisId unter iteration. Eigenheiten (verifiziert): Reservierte/kollidierende Autor-ids (etwa "title") ersetzt der Sanitizer durch Auto-IDs; die Original-id steht NUR im Text-Kanal — bei lossy immer auch content[0].text lesen. | scene.elements zeigt maximal 7 Elemente, suppressed zählt den Rest; Constraints auf verdeckte Elemente werden trotzdem geprüft; einen Nachlade-Parameter gibt es nicht. | PARTIAL = unchecked-Constraints oder vom Cap verdeckte Warnungen (meta.truncated_warnings, nur bei Trunkierung); sichtbare Warnungen allein ergeben PASS. | Farben erscheinen überall als W3C-Namen, nie als Hex; COLOR vergleicht Namen: #ff0000 besteht als red, #ff6347 nicht. | analysisId und Bookmarks leben nur im laufenden Server-Prozess (in-memory) — Ein-Aufruf-Clients können compare/bookmark nicht nutzen. | Liste unvollständig — bei Unerwartetem canvas_validity prüfen, Verlust-Details im Text-Kanal lesen; Glossar: Server-Quickstart. Next step: fix per corrections, then analyze until PASS.

NameTypeReqDescription
constraintsarrayRegeln im Format: "#subject TYPE #reference [value]". Typen: CENTERED-IN, NO-OVERLAP, INSIDE, ALIGNED-LEFT, ALIGNED-TOP, LEFT-OF, ABOVE, DISTANCE-FROM, SAME-SIZE, COLOR
previousIssueCountintegerAnzahl Fehler des vorherigen Aufrufs (fuer Konvergenz-Tracking)
svgstringyesVollstaendiger SVG-String
NameTypeReqDescription
correctionsarrayyes
diffarrayyes
errorobject
iterationobjectyes
metaobject
sceneobjectyes
statusstringyes
uncheckedarrayyes

No examples provided.

vector_mirror_arrange ~279

Computes SVG attributes (x, y, width, height, cx, cy) from canvas dimensions, element definitions, and spatial constraints — layout from scratch, no browser. Input: canvas {width, height}, elements [{id, tag, r?, width?, height?, content?}], constraints (string[]). Output: attributes per element (ready to apply to SVG) + warnings. Eigenheiten (verifiziert): Constraints wirken sequentiell — jede sieht das Ergebnis der vorigen, die Reihenfolge ändert das Layout. | FILL ist arrange-only (in analyze landet es als unchecked); COLOR hat hier keine Wirkung und wird ohne Warnung übersprungen. | Attribute sind Rechen-Vorschläge ohne Canvas-Wächter — auch Positionen außerhalb des Canvas kommen ohne Warnung zurück; mit analyze verifizieren. | Liste unvollständig — nicht Umsetzbares landet als Klartext in warnings; arrange-fähige Typen zeigt vector_mirror_constraints (hasArrange). Next step: use the returned attributes to construct or update the SVG, then call analyze to verify.

NameTypeReqDescription
canvasobjectyesCanvas-Dimensionen
constraintsarrayyesConstraints im Format "#subject TYPE #reference". Reihenfolge ist semantisch relevant.
elementsarrayyesElemente mit ID, Tag und optionalen Dimensionen
NameTypeReqDescription
attributesobjectyes
errorobject
warningsarrayyes

No examples provided.

vector_mirror_bookmark ~231

Saves a previous analyze result as a named baseline for the Sniper-Loop: pin a state, edit the SVG, then compare against the name instead of the UUID. Input: name (kein UUID-Format) + analysisId (UUID aus analyze). Output: name, analysisId, stored, bookmarkCount. Eigenheiten (verifiziert): analysisId und Bookmarks leben nur im laufenden Server-Prozess (in-memory) — Ein-Aufruf-Clients können compare/bookmark nicht nutzen. | Unbekannte/verdrängte analysisId: isError:true und error {code: ANALYSIS_NOT_FOUND, hint}; maximal 10 Bookmarks, das älteste wird verdrängt (LRU). | Liste unvollständig — Session-Begriffe (analysisId, Bookmarks) erklärt das Glossar im Server-Quickstart. Next step: compare(svg, [], name) against this baseline.

NameTypeReqDescription
analysisIdstringyesanalysisId aus vector_mirror_analyze
namestringyesBookmark-Name (kein UUID-Format; Keyspace disjunkt von grids).
NameTypeReqDescription
analysisIdstringyes
bookmarkCountintegeryes
errorobject
namestringyes
storedbooleanyes

No examples provided.

vector_mirror_compare ~336

Compares an SVG with a stored baseline and reports what changed. Input: SVG + analysisId (UUID aus analyze) ODER Bookmark-Name; optionale constraints für Re-Check. Output: analyzeOutput-Form mit diff. Eigenheiten (verifiziert): analysisId und Bookmarks leben nur im laufenden Server-Prozess (in-memory) — Ein-Aufruf-Clients können compare/bookmark nicht nutzen. | Ohne Baseline in diesem Prozess: isError:true und error {code: NO_BASELINE, hint} in structured, derselbe Hint im Text-Kanal — zuerst analyze ausführen. | Diff-Vokabular deutsch und endlich: VERSCHOBEN, FARBÄNDERUNG, FORMÄNDERUNG, NEU, ENTFERNT. | Das Diff sieht Farben auf Namens-Ebene: Hex-Drift unterhalb der Namens-Grenze bleibt unsichtbar, Namens-Sprünge erscheinen als FARBÄNDERUNG. | Liste unvollständig — bei Unerwartetem canvas_validity prüfen, Verlust-Details im Text-Kanal lesen; Glossar: Server-Quickstart. Next step: review diff, adjust SVG, then analyze or compare again.

NameTypeReqDescription
analysisIdstringyesanalysisId (UUID aus analyze) ODER Bookmark-Name (aus vector_mirror_bookmark). Server löst UUID→grids, Name→bookmarks auf. Pflichtfeld (§1.1).
constraintsarrayOptionale Constraints fuer Re-Check
svgstringyesVollstaendiger SVG-String (neuer Zustand)
NameTypeReqDescription
correctionsarrayyes
diffarrayyes
errorobject
iterationobjectyes
metaobject
sceneobjectyes
statusstringyes
uncheckedarrayyes

No examples provided.

vector_mirror_constraints ~166

Lists all available constraint types with their syntax and capabilities. Input: none. Output: types [{type, syntax, hasArrange}]. Eigenheiten (verifiziert): Das Vokabular ist abschließend — was hier nicht steht, existiert nicht; RIGHT-OF/BELOW existieren nicht: LEFT-OF/ABOVE mit getauschten Operanden nutzen. | Unbekannte Typen werden nie geraten: analyze liefert unchecked mit reasonCode CONSTRAINT_TYPE_UNKNOWN und Hint, bei nahen Tippfehlern mit Korrektur-Vorschlag. | Liste unvollständig — unbekannter Typ ⇒ unchecked mit reasonCode + Hint; vollständiges Vokabular: vector_mirror_constraints. Next step: use the constraint syntax in vector_mirror_analyze.

Input schema present but exposes no named parameters.

NameTypeReqDescription
typesarrayyes

No examples provided.

vector_mirror_inspect ~274

Inspects an SVG and returns element positions, sizes, colors and grid mapping without checking constraints. Input: SVG string. Output: scene mit id, tag, Grid-Zelle (Ortssprache, z.B. C4), color je Element. Eigenheiten (verifiziert): Reservierte/kollidierende Autor-ids (etwa "title") ersetzt der Sanitizer durch Auto-IDs; die Original-id steht NUR im Text-Kanal — bei lossy immer auch content[0].text lesen. | Derselbe 7-Elemente-Cap wie analyze — inspect umgeht ihn nicht; suppressed zählt verdeckte Elemente. | structuredContent trägt nur scene (plus meta bei Trunkierung) — kein status, keine corrections; canvas_validity und suppressed stehen in scene. | canvas_validity=lossy heißt: der Sanitizer hat etwas entfernt oder ersetzt — auch render-neutrale Strips (etwa Kommentare) zählen; Verlust-Details stehen nur im Text-Kanal der Antwort. | Liste unvollständig — bei Unerwartetem canvas_validity prüfen, Verlust-Details im Text-Kanal lesen; Glossar: Server-Quickstart. Next step: define constraints based on the layout, then call analyze.

NameTypeReqDescription
svgstringyesVollstaendiger SVG-String
NameTypeReqDescription
errorobject
metaobject
sceneobjectyes

No examples provided.

vector_mirror_palette ~173

Extracts fill and stroke colors from all SVG elements — review the color scheme before applying COLOR constraints. Input: SVG string. Output: colors [{id, fill, stroke}]. Eigenheiten (verifiziert): Farben erscheinen überall als W3C-Namen, nie als Hex; COLOR vergleicht Namen: #ff0000 besteht als red, #ff6347 nicht. | Maximal 7 Einträge in colors, ohne Zähler — weitere Elemente nennt nur die Text-Zeile (N weitere). | Liste unvollständig — bei Unerwartetem canvas_validity prüfen, Verlust-Details im Text-Kanal lesen; Glossar: Server-Quickstart. Next step: use COLOR constraints in analyze to enforce specific colors.

NameTypeReqDescription
svgstringyesVollstaendiger SVG-String
NameTypeReqDescription
colorsarrayyes
errorobject

No examples provided.

vector_mirror_selftest ~220

Runs the 5 calibration fixtures (EK-1 color, EK-2 grid, EK-3 constraint, EK-4 3D-suppression, EK-5 frozen animation) and verifies Vector Mirror against the spec-derived truth. Input: optional full (boolean) — adds an N=10 mini determinism check. Output: status (PASS/FAIL), calibrated, total, failures. Eigenheiten (verifiziert): Misst die 5 Eichkörper gegen spec-abgeleitete expected-Werte, nie gegen gespeicherten eigenen Output — Anti-Zirkularität als Vertrauensanker. | Liste unvollständig — Feld-Bedeutungen erklärt das Glossar im Server-Quickstart; bei selftest-FAIL nennt failures die exakte Spec-Abweichung je Eichkörper. Next step: PASS = trust the measurements; FAIL = fix per failures, then re-run.

NameTypeReqDescription
fullbooleantrue → zusätzlich N=10-Mini-Determinismus-Check (langsamer). Default: nur Kalibrierung.
NameTypeReqDescription
calibratedintegeryes
failuresarrayyes
statusstringyes
totalintegeryes

No examples provided.

vector_mirror_status ~151

Returns Vector Mirror server health. Input: none. Output: version, browser (running/stopped), lastAnalysis, constraintTypes, breaker, calibration. Eigenheiten (verifiziert): calibration=PENDING: der Auto-Selftest vom Serverstart läuft noch oder scheiterte — nicht defekt; der Browser startet lazy beim ersten Mess-Aufruf. | lastAnalysis zeigt Baseline-Existenz nur für diesen Prozess. | Liste unvollständig — Feld-Bedeutungen erklärt das Glossar im Server-Quickstart; bei selftest-FAIL nennt failures die exakte Spec-Abweichung je Eichkörper. Next step: if browser is stopped, any analyze/inspect/palette call will auto-start it.

Input schema present but exposes no named parameters.

NameTypeReqDescription
breakeryes
browserstringyes
calibration
constraintTypesintegeryes
lastAnalysisbooleanyes
versionstringyes

No examples provided.

Common questions

What is the io.github.c64dos-png/vector-mirror MCP server?

io.github.c64dos-png/vector-mirror is an MCP server listed in the public MCP registry as io.github.c64dos-png/vector-mirror. Deterministic SVG perception for LLM agents, measures, never guesses. MCP server. This page covers its npm package (vector-mirror).

Is the io.github.c64dos-png/vector-mirror MCP server safe to use?

io.github.c64dos-png/vector-mirror scores 80 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 20 September 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the io.github.c64dos-png/vector-mirror MCP server expose?

io.github.c64dos-png/vector-mirror exposes 9 tools: vector_mirror_analyze, vector_mirror_compare, vector_mirror_bookmark, vector_mirror_inspect, vector_mirror_palette, and 4 more. Their descriptions and schemas cost roughly 2,254 tokens of context every time the server is loaded.

Is the io.github.c64dos-png/vector-mirror MCP server still maintained?

io.github.c64dos-png/vector-mirror is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What licence is the io.github.c64dos-png/vector-mirror MCP server under?

io.github.c64dos-png/vector-mirror declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.