seedance-continuation

This skill should be used when a Seedance 2.0 user asks to continue, extend, make the next part, repair the tail, bridge between known frames, re-anchor drift, or create a successor prompt from accepted footage.

Quality

83%

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

75%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.

A dense, competent instruction-only skill with strong validation gates, feedback loops, and a concrete output contract, supported by clearly signaled one-level references. It loses points for run-on verbosity, the absence of a worked prompt example, and a workflow that is scattered across sections rather than unified into one sequenced flow.

Suggestions

Add one short worked example showing a completed output contract (continuation type through final Seedance prompt) so the abstract 'translate record into carriers' step is concrete.

Consolidate the high-level procedure (load Director's Read -> classify -> gate -> observe -> compile -> return) into a single numbered 'Workflow' section, then keep the rule sections as supporting detail.

Break the longest compound sentences (especially around the source-gate and authoring-state steps) into shorter steps or a checklist to improve scannability.

Dimension	Reasoning	Score
Conciseness	The body assumes Claude's competence — it never explains what Seedance or a clip is — and every section carries domain-specific procedure Claude would not already know, though several sentences (e.g., the 'After the source gate is satisfied...' paragraph) are dense run-ons that could be tightened.	4 / 5
Actionability	Concrete guidance is present — an exact fallback utterance, a runnable command ('python scripts/extract_last_frame.py <take>'), a named-field input gate, and a numbered output contract — but no full worked example of a compiled continuation prompt is shown, leaving a minor gap.	4 / 5
Workflow Clarity	Validation checkpoints and feedback loops are strong (input gate refusal, intent-echo confirmation 'before generation spends money', repair_tail-before-continue, reanchor_after_drift, max_chain_depth cap), but the workflow is spread across topical sections rather than presented as one clean sequenced procedure.	4 / 5
Progressive Disclosure	References are well-signaled and one level deep with conditional load guidance ('Load failure-atlas when the continuation failed', 'Load directing-engine so the next clip inherits...'), but no references/ bundle is present to verify the files exist, the SKILL.md itself inlines a fair amount of rule content, and eight separate references is a moderately heavy navigation surface.	4 / 5
	Total	16 / 20 Passed

Description

91%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, niche-specific description with an explicit 'Use when' clause and comprehensive natural trigger terms. Its only weakness is that the 'what does this do' capability is entangled with the trigger list rather than stated separately.

Suggestions

Lead with a short capability statement (e.g., 'Generates continuation prompts for Seedance 2.0 clips grounded in accepted footage.') before the 'Use when' trigger list so 'what' and 'when' are both cleanly explicit.

Consider trimming the redundant trio 'continue, extend, make the next part' to the single most natural phrase to avoid trigger-term overlap.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions in the continuation domain — 'continue, extend, make the next part, repair the tail, bridge between known frames, re-anchor drift, or create a successor prompt from accepted footage' — giving comprehensive coverage rather than a single generic verb.	5 / 5
Completeness	The explicit 'should be used when...' clause answers 'when' clearly and the action verbs imply the 'what' (create successor prompts), but the 'what' is folded into the trigger list rather than stated as its own clean capability sentence as in the anchor-5 example.	4 / 5
Trigger Term Quality	Phrases like 'continue', 'extend', 'make the next part', 'repair the tail', 'bridge between known frames', and 're-anchor drift' are natural things a Seedance user would actually say, with synonyms covered and no jargon-only filler.	5 / 5
Distinctiveness Conflict Risk	The scope is tightly bound to 'a Seedance 2.0 user' and continuation-specific triggers, creating a clear niche with minimal overlap risk against unrelated skills.	5 / 5
	Total	19 / 20 Passed

Validation

75%

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 — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
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
relative_links	Relative link issues: 11 suspicious	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	12 / 16 Passed

Repository: Emily2040/seedance-2.0
Path: skills/seedance-continuation/SKILL.md
Commit: 44b5149

Reviewed: 24 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.