plan-canvas

Open plans and HTML artifacts in a local browser canvas where the human annotates elements, chats, and approves or requests changes without leaving the page. Use when presenting a plan for review, or when feedback like "move this, change that" is easier pointed at than typed.

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

88%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.

Highly actionable, executable content with an exceptionally well-instrumented workflow including explicit backstops and failure handling. The main weaknesses are minor rhetorical padding and a monolithic single-file structure that inlines reference-grade detail.

Suggestions

Trim the anthropomorphic framing in the 'Stay listening' and reply sections to plain imperative instructions — e.g. replace 'the human talks to an empty chair' passages with the two or three concrete rules they motivate.

Move the Mermaid rendering details (CDN pinning, ECC_PLAN_CANVAS_MERMAID_URL fallback) and the env-var/state-dir reference (ECC_PLAN_CANVAS_IDLE_MS, ECC_PLAN_CANVAS_STATE_DIR) into a references/ file, keeping SKILL.md to the core loop.

Unify the step numbering — steps 1 and 2 appear only as code comments while steps 3 and 4 are bolded headers; a single numbered list would make the loop sequence scannable at a glance.

Dimension	Reasoning	Score
Conciseness	Commands and examples are tight and efficient, but a few rhetorical passages ("the human talks to an empty chair", "from the human's side of the glass", "indistinguishable from a broken canvas") are colorful over-explanation that could be trimmed.	4 / 5
Actionability	Fully executable copy-paste commands (open, await, --reply, typing --state, end, pending), a concrete example JSON payload, and complete worked flows (plan approval, revision loop) cover the common cases.	5 / 5
Workflow Clarity	The open → await (background) → reply → end sequence is clearly laid out with explicit checkpoints and recovery paths: the pending backstop, the stop:plan-canvas-pending hook, verdict/ended status handling, re-run guidance, and offline Mermaid degradation.	5 / 5
Progressive Disclosure	Well-organized single-file skill with clear sections and one clearly signaled one-level design-doc link, but at ~190 lines with no bundle split, detail such as Mermaid rendering internals and env-var reference material sits inline rather than in a separate reference file.	4 / 5
	Total	18 / 20 Passed

Description

92%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: concrete multi-action capability statement, an explicit 'Use when…' clause with natural quoted trigger phrases, and a distinctive niche. Only minor trigger-synonym gaps keep trigger term quality below full marks.

Dimension	Reasoning	Score
Specificity	"Open plans and HTML artifacts in a local browser canvas where the human annotates elements, chats, and approves or requests changes" lists multiple specific concrete actions (open, annotate, chat, approve/request changes) comprehensively covering the skill's capabilities.	5 / 5
Completeness	Explicitly answers both what (first sentence describes the full mechanism) and when ("Use when presenting a plan for review, or when feedback like 'move this, change that' is easier pointed at than typed") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural phrases like "presenting a plan for review", "move this, change that", and "easier pointed at than typed" are present alongside "browser" and "HTML", but a few natural synonyms users might say (e.g. "visual review", "annotate", "design feedback") are missing.	4 / 5
Distinctiveness Conflict Risk	The local-browser-canvas annotation/approval mechanism is a clear niche with distinct plan-review triggers, giving minimal overlap risk with code-review or generic browser-preview skills.	5 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: affaan-m/ECC
Path: skills/plan-canvas/SKILL.md
Commit: c70874f

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.