BC sn-ppt-standard
Standard and fast PPT pipeline. All LLM / VLM / T2I calls are wrapped in a single CLI entry (scripts/run_stage.py). The main agent's job is simple: emit ONE shell command per stage, never write loops, never write prompts. Standard mode plans thoroughly with a three-sample deck preview checkpoint (three concatenated deck images plus a preview URL), web research, image search, and user-selected final output format (PPTX or PDF) for polished, delivery-ready presentations. Fast mode builds a complete draft immediately with autonomous decisions, then provides structured refinement suggestions so the user can iterate quickly. Supports AI-generated infographics (U1) for diagrams and flowcharts, web image search (Serper) for real photos, and ECharts for data charts.
Standard and fast PPT pipeline.
As a process C 64/100 · Has gaps — weak spots: inputs and preconditions, execution cost
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
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
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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
✓ No critical or high findings
Medium and low: 3
-
medium Exfiltration
net-redirectable-api-keyscripts/launch_workbench.py:377Helper sends the API key to a host configured by an environment variable — the key can be redirected to another serverAPI key + configurable base URL from environment
-
low Secrets in code
secret-high-entropy-tokenscripts/export_pptx/package-lock.json:44High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)"integrity": "sha5…uWm+fFRcIOgKBMiOBP+eXiy…9ab+DDKA==",
detector -
low Secrets in code
secret-high-entropy-tokenscripts/export_pptx/package-lock.json:181High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)"integrity": "sha5…pjs+EVnW…hE7+ZyY8…tdA==",
detector
Files scanned: 37. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 8994 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "triggers"
Process rating: all ten parameters 64/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 40Execution cost. Instruction body is 8994 tokens: crowds the task out of the window
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, web, python, node) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 85Steps. 107 steps, 2 vague phrases
- 100Failures and branches. 5 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 15 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (19 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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- -5TODO / placeholder text left in the skill
- -2localhost URLs: will not work for another user
- -31 of 6 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 769: enough signal without eating the budget
- +4Structure: 27 headings
- +3Step-by-step instructions: 107 items
- +3Output format is stated explicitly
- +4Has examples (7 code blocks)
- +4Reference files are cited in the instructions (1 of 2)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 58.