DD ofox-video-core
Requires OFOX_API_KEY — create one at https://app.ofox.ai. Shared execution layer for the Ofox video generation API (api.ofox.ai) — creates a video job, polls it to completion, downloads the finished mp4 from a persistent CDN URL, and reports the real cost. This is a library skill, not a standalone user-facing one — it is invoked by scenario skills such as seedance-short-drama, seedance-ad-creative, and seedance-product-video, which build model/prompt/resolution choices for a specific use case and then call into this skill's script rather than re-implementing the API calls. Load this skill directly only when a user explicitly names the Ofox video API, asks to call it with specific low-level parameters, or asks to debug/resume a stuck or failed Ofox video job by job id — for a plain scenario request ("make me a short drama scene", "generate a cinematic ad clip"), use the relevant scenario skill instead, which itself depends on this one.
Requires OFOXAPIKEY — create one at https://app.ofox.ai. Shared execution layer for the Ofox video generation API (api.ofox.ai) — creates a video job, polls…
As a process D 41/100 · Unfinished process — References files that are not bundled: ../previs-rerender/SKILL.md
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
The skill tells the agent to hide things from you: not to show errors, not to mention actions, to report differently from what was done. You lose the ability to see what the agent really did.
Transparency beats a smooth answer. If the goal is to hide technical noise, ask the agent to "summarise briefly", not to "not mention".
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
How to improve
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- The text references files that are not there: add them or drop the references.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 3
-
high Concealment
en-hide-from-userreferences/prompt-structure.md:951Instruction to hide actions from the usernot merely crop the picture — it can silently delete part of what a
Medium and low: 2
-
medium Exfiltration
net-credential-useSKILL.md:992Credential used in a network call (verify the destination is the intended service)curl -s -H "Authorization: Bearer $OFOX_API_KEY" \
-
low Exfiltration
net-credential-usereferences/ofox-video.sh:1568Credential used in a network call (verify the destination is the intended service) (quoted — discussed, not commanded)echo " curl -s -H \"Authorization: Bearer \$OFOX_API_KEY\" \"${API_BASE}/videos?limit=5\" | jq ." >&2quoted
Files scanned: 27. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 15278 tokens (recommended < 5000); move details to references/ - warning
missing-refreference to a missing file: ../previs-rerender/SKILL.md - note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 41/100
- 0Tools and files. 1 referenced file(s) missing: ../previs-rerender/SKILL.md
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 50 mutating operations with no state check
- 40Execution cost. Instruction body is 15278 tokens: crowds the task out of the window
- 100Steps. 27 steps
- 100Failures and branches. 5 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 17 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (7 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 949: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +4Structure: 38 headings
- +3Step-by-step instructions: 27 items
- +4Has examples (23 code blocks)
- +4Reference files are cited in the instructions (5 of 5)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 74.