CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

real-time-collab

Multi-user collaborative editing with Yjs CRDT, SSE fast-path transport, and granular server-side merge. Use when adding real-time collaborative editing to a template, debugging sync issues, or understanding how the agent and humans edit documents simultaneously.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/real-time-collab/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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 highly actionable and framework-specific with a clean numbered setup workflow, concrete code for every integration point, and an excellent pitfalls section covering real failure modes. Its weaknesses are length and monolithic structure: duplicated examples, very dense run-on paragraphs, and deep-dive topics (Live Copies, poll-boost timing, saved copies) inlined in SKILL.md instead of split into reference files.

Suggestions

Move the Live Copies contract, the collab poll boost timing internals, and the "Opening from a saved copy" behavior into one-level-deep reference files (e.g., references/live-copies.md, references/sync-cadence.md), leaving a short summary and a clearly signaled link in SKILL.md.

Deduplicate the createCollabPlugin resource-mode example (shown verbatim in Security and again in Enabling step 2) and trim the "Missing access" pitfall to a one-line cross-reference to the Security section.

Break the multi-clause run-on bullets (e.g., the ~15-line "Collab poll boost" bullet and the Live Copies prose) into shorter scannable statements, keeping only timing constants and rules.

Add an explicit verification step to the setup workflow (e.g., check the Doctor output or confirm /_agent-native/collab/:docId/state returns after wiring) so the sequence ends in a checkpoint.

DimensionReasoningScore

Conciseness

The body assumes Claude knows Yjs/CRDT/SSE (never explains them), so it is above anchor 2, but it is still ~520 lines that could be tightened: the createCollabPlugin resource-mode example appears verbatim in both Security and Enabling step 2, the "Missing access" pitfall re-states the Security section, and dense run-on passages like the 15-line "Collab poll boost" timing bullet pad the token budget without setup-flow value.

3 / 5

Actionability

The five-step "Enabling Collaboration" flow (pnpm add, server plugin, useCollaborativeDoc, TipTap extensions, optimizeDeps) is copy-paste ready and covers the common case, but several advanced snippets are illustrative fragments — "resolveRect={(edit) => /* map descriptor → DOMRect, or null */ null}", "apply: (ops) => applyGranularOps(ops)" — leaving minor gaps versus the fully-executable anchor 5.

4 / 5

Workflow Clarity

"Enabling Collaboration" is a clearly numbered 1–5 sequence, and pitfalls supply error-recovery guidance ("false means delivery failed, so retry later instead of saving"), but verification is implicit — Doctor is only mentioned as flagging misconfigurations, with no explicit checkpoint after setup — which is a minor validation gap versus anchor 5.

4 / 5

Progressive Disclosure

Section headers and the routes table give real structure, but the entire 23 KB skill is one inlined file: the "Live Copies" deep-dive, the "Collab poll boost" timing internals, and the ~40-line "Opening from a saved copy" subsection clearly belong in one-level-deep reference files, and no references/, scripts/, or assets/ bundle exists.

3 / 5

Total

14

/

20

Passed

Description

83%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.

A strong description: it states three concrete, framework-specific capabilities, uses third-person voice, and pairs them with an explicit 'Use when...' clause containing three concrete trigger scenarios. The only gaps are missing common synonym variations (e.g., "collab", "multiplayer") and slight overlap risk with the real-time-sync skill on the sync-debugging trigger.

DimensionReasoningScore

Specificity

Names several concrete capabilities — "Multi-user collaborative editing with Yjs CRDT, SSE fast-path transport, and granular server-side merge" — plus actions via the trigger clause ("adding", "debugging", "understanding"). It lists several specific items but they are technologies rather than a comprehensive action list, so it sits below anchor 5 and above the 1–2-action level of anchor 3.

4 / 5

Completeness

Explicitly answers both: the "what" is the three named capabilities, and the "when" is a concrete "Use when adding real-time collaborative editing to a template, debugging sync issues, or understanding how the agent and humans edit documents simultaneously" clause with three distinct trigger phrases, matching the anchor-5 example structure; anchor 4 would require the when-clause to be less specific.

5 / 5

Trigger Term Quality

Natural phrases users would say are present — "real-time collaborative editing", "debugging sync issues", "edit documents simultaneously" — but common variations like "collab", "multiplayer editing", or "live cursors" are missing, so it is good-but-not-comprehensive coverage rather than the synonym breadth of anchor 5.

4 / 5

Distinctiveness Conflict Risk

The niche (CRDT-based multi-user editing in this framework) is distinct with targeted triggers, but "debugging sync issues" could overlap with the closely related real-time-sync skill referenced in Related Skills, giving minor overlap risk rather than the minimal-conflict profile of anchor 5.

4 / 5

Total

17

/

20

Passed

Validation

75%

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

Validation — 12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (522 lines); consider splitting into references/ and linking

Warning

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

12

/

16

Passed

Repository

BuilderIO/agent-native

Path

.agents/skills/real-time-collab/SKILL.md

Reviewed

Table of Contents

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.