Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add mrDesign-ww/vault-os --skill seedance-troubleshootgit clone --depth 1 https://github.com/mrDesign-ww/vault-osWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mrdesign-ww/vault-os/seedance-troubleshoot)<a href="https://agentmods.dev/skills/mrdesign-ww/vault-os/seedance-troubleshoot"><img src="https://agentmods.dev/badge/skills/mrdesign-ww/vault-os/seedance-troubleshoot/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mrdesign-ww/vault-os/seedance-troubleshoot"><img src="https://agentmods.dev/badge/skills/mrdesign-ww/vault-os/seedance-troubleshoot.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00056 | $0.01324 |
| Opus 5 | $0.00028 | $0.00662 |
| Sonnet 5 | $0.00011 | $0.00265 |
| Haiku 4.5 | $0.00006 | $0.00132 |
Grade A, and why
seedance-troubleshoot scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
88% identical to seedance-troubleshoot — 16 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
seedance-troubleshoot
If the take is partially good rather than failed, route to [ref:retake-protocol] triage first - most takes deserve a verdict, not a rewrite. Diagnose failure before rewriting. Do not simply add more adjectives. Identify whether the failure came from mode mismatch, overload, ambiguity, fragile identity, unsafe wording, unsupported platform behavior, or missing preservation constraints.
When the diagnostic tree has no row for the failure, load [ref:model-mechanics] and diagnose by mechanism: attention dilution, prior conflict, summoned negation, broken trajectory, compounding error, conditioning conflict, capacity starvation, or an overloaded audio-video joint constraint. Load [ref:field-observed-tips], [ref:reference-workflow], and [ref:api-workflow] when the failure involves continuation, edit/extend, source clips, audio references, or platform-specific errors. Load [ref:shot-list-continuity] for multi-shot drift and [ref:delivery-qc] for final-client or delivery failures. When sequence state is present, load [ref:failure-atlas], [ref:continuation-handoff], and [ref:continuity-qc]; diagnose against continuity locks, completed beats, exact reference tags, and reserved future beats.
Intent
A failed generation feels personal - the user showed the machine their idea and the machine returned something broken. The soul of this skill is rescue without blame: name the mechanism, never the user; save the idea, not just the prompt. They should leave with a fix and their confidence intact.
Diagnostic Tree
| Symptom | Likely cause | First repair |
|---|---|---|
| Product or face changes | I2V prompt re-described visible identity or overloaded motion. | Add preservation constraints; remove duplicate static detail. |
| Camera jumps | Several incompatible moves or no endpoint. | Choose one move with start and finish. |
| Generic output | Hollow style words and weak action. | Replace with physical action, source light, material, and sound. |
| Motion ignored | Static prompt or no visible consequence. | Add actor, verb, timing, and changed end state. |
| Lip-sync poor | Moving head/camera, long dialogue, unassigned speaker. | Lock framing, shorten line, assign speaker. |
| VFX noisy | Effect has no source, physics, or dissipation. | Add source, material, path, interaction, and endpoint. |
| Prompt blocked | Protected IP, real-person, graphic, or bypass-like wording. | Rewrite intent in safe production language without evasion. |
| Extension quality degrades | No last-frame anchor or too many new variables across continuations. | Use returned last frame as first frame and change one variable. |
| Audio reference ignored | Competing video sound, no visual beat mapping, or unsupported combo. | Mute competing video and map one visible event to the beat. |
| Text/logos break | Small text asked to move or be redrawn. | Keep text static, centered, and protected; animate light around it. |
| Client QC fails | Prompt output treated as final delivery without post/QC. | Route to delivery preflight, post fix, or regenerate only the failing shot. |
| Continuation assumed planned ending | Previous clip was not reviewed or observed_end_state was ignored. | Replace opening with actual observed end state. |
| Previous action restarted | Completed beat was not marked already_happened. | Add completed beat exclusion. |
| Future beat leaked | Reserved beat entered the current prompt. | Remove future beat and stop earlier. |
| Identity reference conflicts with continuity source | Source clip controlled transient state and identity at once. | Re-anchor identity from canonical reference. |
| Screen direction reset | Axis relation was not locked or intentionally reset. | Preserve screen direction or declare new-shot axis reset. |
| Open motion lost | Subject or camera vector was not inherited. | Carry motion vector into the opening sentence. |
| Camera phase restarted | Parent camera endpoint was not recorded. | Start from observed camera phase. |
| Prop state contradicted | Owner, position, or condition was missing. | Add prop state handoff. |
| Audio phase restarted | Completed dialogue or music phase was not logged. | Continue or intentionally change the audio phase. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 73 lines · 56 tokens per session scan A 9829ea8901f6
seedance-troubleshoot is a skill published in the GitHub repository mrDesign-ww/vault-os (2 stars, last pushed 3d ago), licensed MIT. It adds 56 tokens to every session and 1,324 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to seedance-troubleshoot, differing in 16 lines, and is treated as a copy.
Other skills, from other repositories
seedance-troubleshoot
This skill should be used when a Seedance 2.0 output is blurry, jittery, off-prompt, morphing, blocked, visually generic, unstable, desynced, inconsistent, or otherwise fails and needs root-cause diagnosis.
seedance-troubleshoot
This skill should be used when a Seedance 2.0 output is blurry, jittery, off-prompt, morphing, blocked, visually generic, unstable, desynced, inconsistent, or otherwise fails and needs root-cause diagnosis.
seedance-troubleshoot
This skill should be used when a Seedance 2.0 output is blurry, jittery, off-prompt, morphing, blocked, visually generic, unstable, desynced, inconsistent, or otherwise fails and needs root-cause diagnosis.
huawei-cloud-mrs-host-fault-diagnose
Huawei Cloud MRS cluster fault diagnosis skill. Diagnoses service faults, instance faults, and host faults through progressive root cause localization: quick log scan first, host troubleshooting when host issues are found, detailed investigation when no conclusion is reached. Driven by the built-in LakeWatch API…
huawei-cloud-mrs-host-alarm-diagnose
Huawei Cloud MRS cluster alarm diagnosis skill. Analyzes the root cause of an MRS alarm based on user-provided alarm information (alarm ID, alarm name, alarm details, occurrence time, node IP, related service and logs), then outputs the root cause, repair steps, and verification method. Diagnosis is driven by the…
lottie-motion-builder
Build or refine small production-ready Lottie animations from SVGs, logos, UI states, loaders, and branded motion assets with a local preview harness, explicit inputs, and output verification. Use when the user asks for Lottie, JSON animation, SVG reveal animation, logo animation, or lightweight motion assets for…