DD capturas-tutorial-formula100k
Genera CAPTURAS UI tipo tutorial (cards de paso numerado, code blocks tipo terminal macOS con botón Copiar, callouts info/warning/success, screenshot frames, atajos de teclado, links, comparison antes/después, listas) con la estética F100K (fondo crema, círculos coral, DM Serif Display, Inter, JetBrains Mono). Render HTML+CSS+Puppeteer → PNG transparente o crema, 1920×1080. NO usa Nanobanana ni Higgsfield (la IA rompe el texto; HTML es pixel-perfect). Activar cuando pidan "diséñame una captura tipo tutorial", "imagen del comando", "renderiza un terminal con este comando", "card de paso 1/2/3", "callout warning/info/success", "tarjeta de atajo Cmd+V", "comparison antes/después", "imagen UI para mi landing/artifact/carrusel", "overlays UI para mi video tutorial", "convierte este guion en capturas tutorial". 2 modos: STANDALONE (PNGs sueltos) y BATCH (MANIFEST_UI.yml). También la invoca recursos-de-video-formula100k. NO renderiza video (editor-video-formula100k). NO usa stock.
Genera CAPTURAS UI tipo tutorial (cards de paso numerado, code blocks tipo terminal macOS con botón Copiar, callouts info/warning/success, screenshot frames…
As a process D 36/100 · Unfinished process — References files that are not bundled: templates/_tokens.css
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 skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
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.
- 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 · 5
✓ No critical or high findings
Medium and low: 5
-
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bashallowed-tools: Bash Read Write Edit AskUserQuestion
-
low Dangerous commands
cmd-pipe-to-shell-known-hostscripts/install.sh:50Pipe-to-shell installer from a well-known host (still executes remote code) (string literal in code, not executed)echo " node $SKILL_DIR/scripts/render.mjs --type terminal --out /tmp/test.png --data '{\"command\":\"curl -fsSL https://claude.ai/install.sh | bash\"}'"code literal -
low Dangerous commands
cmd-pipe-to-shell-known-hostscripts/render.mjs:295Pipe-to-shell installer from a well-known host (still executes remote code) (string literal in code, not executed)--data '{"command":"curl -fsSL https://claude.ai/install.sh | bash"}'code literal -
low Dangerous commands
cmd-pipe-to-shell-known-hostSKILL.md:138Pipe-to-shell installer from a well-known host (still executes remote code) (quoted — discussed, not commanded)--data '{"command":"curl -fsSL https://claude.ai/install.sh | bash"}'quoted -
low Dangerous commands
cmd-pipe-to-shell-known-hostSKILL.md:203Pipe-to-shell installer from a well-known host (still executes remote code) (quoted — discussed, not commanded)command: "curl -fsSL https://claude.ai/install.sh | bash"
quoted
Files scanned: 13. 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 ≈ 5250 tokens (recommended < 5000); move details to references/ - warning
missing-refreference to a missing file: templates/_tokens.css
Process rating: all ten parameters 36/100
- 0Tools and files. 1 referenced file(s) missing: templates/_tokens.css
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 50Failures and branches. 0 branches, has a failure section
- 70Execution cost. Instruction body is 5250 tokens
- 85Steps. 72 steps, 3 vague phrases
- 100Consistency. Name and required fields are in place
- 100Running it twice. No mutating operations
- low 14 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (9 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 988: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -5TODO / placeholder text left in the skill
- -217 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 10 example trigger phrases
- +4Structure: 30 headings
- +3Step-by-step instructions: 72 items
- +4Has examples (6 code blocks)
- +3All 2 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 54.