keyline
OCI · GHCR.IO/KEYLINE-DEV/KEYLINE-MCP · 2 COMPONENTS
Design engine for AI agents: images at every size, no Chrome, 2× fewer tokens
Available components
Last scanned (15 hours ago)
Scores higher than 57% of OCI servers
Category Design and Media
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 Security0
- No malware scan is available for this kind of package: the supply-chain vendors we use do not cover it. This is a permanent gap in our coverage, not a finding about the package.Unverified
- Known CVEs could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.Unverified
- Install-script risk not yet assessed.Unverified
- Dependency health could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.Unverified
Provenance & Transparency35
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- License check failed: the license (FSL-1.1-ALv2) isn't a recognized OSI-approved license. See how to fix → Fail
- Actively maintained (last published 40 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability54
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 2488 tokens (~414/item across 6 items; 6 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 Management100
- No destabilizing schema changes in the last 30 days.Pass
Tool Coverage88
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 65% of tool parameters carry a description.Partial
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 6 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 7 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a current MCP spec version (2026-07-28).Pass
Unverified: 1 category
A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.
Checks we ran (12)
- Source repository
- Provenance
- License
- Maintenance
- Security disclosure
- Schema quality
- Stability
- Tool coverage
- Injection markers
- Destructive annotations
- Judged manipulation
- MCP protocol
Checks we could not complete (4)
These are gaps on our side, not findings about this package. We only credit what we can confirm, so a category made up entirely of them scores 0.
- Malware scan in Supply Chain Security (scores 0 for this)No malware scan is available for this kind of package: the supply-chain vendors we use do not cover it. This is a permanent gap in our coverage, not a finding about the package.
- Known CVEs in Supply Chain Security (scores 0 for this)Known CVEs could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.
- Install scripts in Supply Chain Security (scores 0 for this)Install-script risk not yet assessed.
- Dependency health in Supply Chain Security (scores 0 for this)Dependency health could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.
Does not apply to this package (2)
- Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
- Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
The score does not measure everything. What the score does not measure →
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.
- 13 Nov 26 +1
- Stability: 0.97 → pass security
- 11 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 10 Nov 26 +5
- Source repository: unverified → pass ▲ security
- 9 Nov 26 −7
- Provenance: pass → unverified ▼ security
- Stability: pass → 0.87 functional
- 5 Nov 26 +5
- Stability: fail → pass ▲ security
- 31 Oct 26 0
- Provenance: unverified → fail ▼ security
- 30 Oct 26 0
- Provenance: fail → unverified ▼ security
- 29 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 60 to 63.
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 21 Nov 2026 · Analysed oci/ghcr.io/keyline-dev/keyline-mcp@0.8.1
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | oci |
| Reason | No attestation published |
Background: How many MCP packages publish verified provenance →
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 →
asset_add ~139
Add an image (PNG, JPEG, SVG) from url, path or base64. Returns id and size. Or a video (mp4, mov, webm) or sound (mp3, m4a, wav) by path or url.
| Name | Type | Req | Description |
|---|---|---|---|
| base64 | string | – | Or the file bytes, base64: small images only (a logo); a photo goes by url or path. |
| id | string | – | Id for layers; generated if omitted. |
| path | string | – | Or a local file in ~/keyline |
| sceneId | string | yes | – |
| url | string | – | Public http(s) URL. |
No output schema declared.
No examples provided.
layer_add ~1,292
Add layers, drawn bottom to top. CSS names and values. Any layer: {type, id?, role?, parent? (frame id), x, y (px or "25%"), width, height (px, "hug", "fill", "40%"), min/maxWidth, min/maxHeight, aspectRatio, place top-left…bottom-right + margin, constraints {horizontal: left|right|center|stretch|scale, vertical: top|bottom|center|stretch|scale}, hidden, opacity, rotate (deg), blendMode, style (name or list), media {size id or landscape|square|portrait|wide|tall: {fields}}, fill, stroke, shadow (each one or a list), blur, backdropBlur, borderRadius (px, [tl,tr,br,bl], "full"), mask, scale, translate [x,y], skew, flipX, flipY, edges {sides, depth, seed}}. frame: children, clipsContent (true). Stack: flexDirection row|column|row-reverse|column-reverse (a list is tried in order), gap, padding, alignItems stretch (default)|flex-start|center|flex-end|baseline, justifyContent flex-start|center|flex-end|space-between|space-around|space-evenly, flexWrap; children: alignSelf, flexGrow (alone, it fills), layoutPriority (number; lower gives way first), position absolute. Grid: gridTemplateColumns/Rows (CSS tracks, repeat(auto-fill, minmax(160px, 1fr))), gridTemplateAreas ["hero side", …], gap, padding; children: gridArea, or gridRow/gridColumn ("1 / span 2"). spacer: free space in a stack. firstFit: children; draws the first that fits. text: text (markup <b> <i> <u> <s> <sup> <sub> <br>, <span style="color:…;font-weight:…">, style names as tags), fontSize (the max when fitting), fontWeight, color, fontFamily (Inter or any Google Font), textAlign, textAlignVertical, lineHeight (× fontSize), letterSpacing (px), fill (paints the letters, wins over color), textTransform, fontStyle, textDecoration, maxLines, textWrap balance|pretty, trim "cap", highlight, padding, curve, leader ("." fills a tab gap), knockout. Text box: width+height → shrinks to fit (minimumScaleFactor 0.5) then ellipsis; width → wraps down; neither → one line. image: asset, fit cover|contain|fill|tile, focus [x,y…
| Name | Type | Req | Description |
|---|---|---|---|
| components | object | – | Named layer trees to add or replace, placed by `use` layers. |
| layers | array | yes | – |
| sceneId | string | yes | – |
| styles | object | – | Named styles to add or replace: layer fields a layer's `style` pulls in. |
| tokens | object | – | Named values to add or replace, used as {{name}} in any field or text. |
No output schema declared.
No examples provided.
layer_update ~172
Change or delete layers atomically. Each op: {target: {id}|{role}|{style}|{component, role?}, set: {fields}} or {target, delete: true} or {target: {id}, detach: true} (a use becomes plain layers). A role targets every layer with it; a style or component target changes it everywhere it's used. tokens {name: value} changes tokens. null resets a field. {target: {scene: true}, set: {background, sizes, duration, fps, loop, audio {asset, volume, trimStart, fadeIn, fadeOut}}} changes the scene.
| Name | Type | Req | Description |
|---|---|---|---|
| ops | array | yes | – |
| sceneId | string | yes | – |
| tokens | object | – | Tokens to change; every field using one follows. |
No output schema declared.
No examples provided.
render ~213
Render each size (PNG, or format jpeg|webp|pdf). Returns each size's path and layout as drawn; maxKB lowers quality to fit; rows renders a variant per row of tokens. preview adds a small image of all sizes. apng, gif, mp4 or webm (mp4/webm need ffmpeg) renders the motion, its preview key moments; time (s) a still; muted drops the sound.
| Name | Type | Req | Description |
|---|---|---|---|
| format | string | – | png|jpeg|webp|pdf|apng|gif|mp4|webm |
| maxKB | integer | – | – |
| muted | boolean | – | – |
| preview | boolean | – | – |
| quality | integer | – | 0–100 (90) |
| rows | array | – | Variants: one render per row of token values, e.g. [{"headline": "Sale"}] |
| sceneId | string | yes | – |
| sizes | array | – | Size ids; all if omitted. |
| time | number | – | – |
No output schema declared.
No examples provided.
scene_create ~250
Create a scene from sizes, or from a template by url or path with its tokens set. Returns its id.
| Name | Type | Req | Description |
|---|---|---|---|
| background | string | – | Any CSS color (#FFFFFF). |
| duration | number | – | Seconds: makes it move |
| fps | number | – | (30) |
| height | number | – | – |
| loop | boolean | – | – |
| name | string | – | Id and folder (summer-sale) |
| path | string | – | or a local file in ~/keyline |
| sizes | array | – | Target sizes: {id, width, height, scale, safeArea}, a preset (instagram-portrait|-square|-story, facebook-feed, linkedin-post, x-post, youtube-thumbnail, iab-medium-rectangle|-leaderboard|-skyscraper… |
| tokens | object | – | The template's variables to set: {name: value} |
| url | string | – | Or start from a template: a scene JSON file by url |
| width | number | – | Master size, px; the first size's when omitted. |
No output schema declared.
No examples provided.
scene_describe ~110
Problems per size, or ok. Defects: !truncated|!overflow (W×H) !clipped !hidden !overlaps !covered !unsafe (safeArea). Advisory: warn contrast. full: the layout.
| Name | Type | Req | Description |
|---|---|---|---|
| full | boolean | – | Every layer's box, not just warnings. |
| sceneId | string | yes | – |
| size | string | – | One size id; all sizes if omitted. |
| view | string | – | Image id: the photo, with px rulers. |
No output schema declared.
No examples provided.
How do I install the keyline MCP server?
keyline runs locally as a container image, launched with docker run --rm -i ghcr.io/keyline-dev/keyline-mcp:0.8.1. The registry lists runner options for this package that are not in this command (they are not ones we reproduce, or they name another package). Check the registry entry before relying on it. Ready-made configuration for Claude, Cursor, VS Code, Codex and 3 more is on this page, copied from each client's own documentation.
oci · ghcr.io/keyline-dev/keyline-mcp
claude mcp add keyline-dev-keyline -- docker run --rm -i ghcr.io/keyline-dev/keyline-mcp:0.8.1
The registry lists runner options for this package that are not in this command (they are not ones we reproduce, or they name another package). Check the registry entry before relying on it.
What is the keyline MCP server?
keyline is an MCP server listed in the public MCP registry as io.github.keyline-dev/keyline. Design engine for AI agents: images at every size, no Chrome, 2× fewer tokens. This page covers its container image (ghcr.io/keyline-dev/keyline-mcp).
Is the keyline MCP server safe to use?
keyline scores 48 out of 100 on VerifyMCP. 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 keyline MCP server expose?
keyline exposes 6 tools: asset_add, layer_add, layer_update, render, scene_create, scene_describe. Their descriptions and schemas cost roughly 2,176 tokens of context every time the server is loaded.
Is the keyline MCP server still maintained?
keyline is still listed as active in the MCP registry. We last reached this channel on 21 November 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 keyline MCP server under?
keyline declares the FSL-1.1-ALv2 licence, which is not on the OSI-approved list. Read the terms before using it at work, and note this covers the source only, not the cost of any service it calls.
What did VerifyMCP test on the keyline MCP server?
VerifyMCP ran 12 checks on keyline, covering Source repository, Provenance, License, Maintenance and 8 more. 4 items could not be completed, a gap on our side rather than a finding about keyline: Malware scan, Known CVEs, Install scripts and Dependency health. One category scores 0 for that reason, because we only credit what we can confirm. Endpoint Security and Transport & Reachability do not apply to a container image, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.