tldraw-offline

Drive and script tldraw offline canvases with an agent.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./optional-skills/creative/tldraw-offline/SKILL.md

Quality

Content

77%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an unusually actionable, verification-first manual: exact commands, real bundle scripts, explicit success criteria, and hard-won pitfalls that assume Claude's competence. Its two weaknesses are inline bulk (API contract, prop schema, cua-driver notes) that belongs in references/ files, and a few padded, version- or host-specific narratives that could be trimmed or split out.

Suggestions

Move the cua-driver/foreground-delivery narrative, the inotify host-limit notes, and other version-pinned details into a separate reference file (e.g. references/computer-use-testing.md) and keep a one-line pointer in SKILL.md.

Move the full shape-props schema table and the ctx.editor/ctx.helpers API listing into references/ files, leaving only the prop-enum essentials and the script contract inline.

Tighten the Interactive UI section: the verification-boundary prose can be reduced to the two caveats as bullets, dropping narrative details like 'Verified by the Cua team on this exact v1.11.0 AppImage'.

Dimension	Reasoning	Score
Conciseness	Most of the body is dense, non-obvious operational knowledge (per-call token reads, listener-fires-after-commit, required richText), but the cua-driver foreground-delivery narrative, the inotify host-limit details, and the version-pinned claim 'Verified by the Cua team on this exact v1.11.0 AppImage' go beyond minor over-explanation and time-sensitive detail is not isolated in a deprecated/old-patterns section — matching 'mostly efficient but includes some unnecessary explanation or could be tightened'.	3 / 5
Actionability	The skill gives copy-paste-ready curl commands with exact endpoints, headers, and inline token extraction; an executable document-script contract; a concrete hit-testing pattern; prop tables with validated required fields; and real, matching bundle files (scripts/main.js, scripts/counter.js, scripts/validate_shapes.mjs) — fully executable with specific examples covering the common cases.	5 / 5
Workflow Clarity	Two workflows are explicitly selected by criterion (must the change survive reload?), followed by a 6-step Procedure and a dedicated Verification section with explicit success criteria (script-status digest equality, 'state: applied') and error-recovery paths (retry pending, read errorLogPath, report instead of claiming success) — clear sequencing with validation feedback loops.	5 / 5
Progressive Disclosure	The body is well-sectioned and its three bundle scripts are clearly signaled and verified to exist one level deep, but ~270 lines of content that clearly belongs in separate reference files — the full script API contract, the shape-props schema table, and the computer-use/cua-driver notes — are inlined in SKILL.md, matching 'content that should be separate is inline' rather than level 4's 'most content is appropriately placed'.	3 / 5
	Total	16 / 20 Passed

Description

57%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is short, third-person, and highly distinctive, naming the exact product and artifact. Its weakness is completeness and trigger coverage: it lacks any explicit 'Use when...' trigger clause and omits natural synonyms (whiteboard, diagram, drawing, .tldraw) that would help a user or Claude select it.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user has tldraw offline open or asks to build, edit, or script a tldraw canvas/whiteboard/diagram, or to add interactive behavior to a .tldraw file.'

Include natural synonyms users actually say — 'whiteboard', 'diagram', 'drawing', '.tldraw' — alongside 'canvases' to improve trigger matching.

Make the capabilities slightly more concrete (e.g. 'read and edit the canvas via its local HTTP API; write embedded document scripts') instead of the generic verbs 'drive and script'.

Dimension	Reasoning	Score
Specificity	"Drive and script tldraw offline canvases" names the domain plus two actions ("drive", "script"), matching the 'names domain and 1-2 concrete actions' anchor; it does not list several specific concrete actions like the level-4/5 anchors, and it is more concrete than the minimal/generic level-2 anchor.	3 / 5
Completeness	The 'what' is reasonably clear (drive and script tldraw offline canvases), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines — squarely the 'clear what, when missing' anchor.	3 / 5
Trigger Term Quality	The description carries the core natural terms "tldraw", "offline", "canvases", and "agent", but misses common variations and synonyms a user would say — "whiteboard", "diagram", "drawing", and the ".tldraw" file extension — so it fits 'some relevant keywords but missing common variations or synonyms' rather than level 4's 'a few natural terms missing'.	3 / 5
Distinctiveness Conflict Risk	"tldraw offline canvases" is a clear niche naming a specific product and artifact type with distinct triggers, giving minimal overlap risk with other skills; nothing generic that would fire on the wrong skill.	5 / 5
	Total	14 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: NousResearch/hermes-agent
Path: optional-skills/creative/tldraw-offline/SKILL.md
Commit: 68fa7e9

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.