AC deck-pipeline
Production-grade Claude Code system that takes presentation decks from raw Chinese draft to McKinsey-polished English — with full audit trail, layout integrity checks, and a swappable PROFILE block for project-specific defaults. Built on a 4-stage pipeline (Sense Pass → McKinsey Translation → Layout Audit → Handoff). Also runs polish-only on any single-language deck. TRIGGER when the user: • hands over a .pptx containing Chinese and asks for English / translation • asks for "deck pipeline", "deck polish", "deck globalizer" • asks for layout polish, font cleanup, overflow fixing on any deck • asks to update / reverse-sync a bilingual comparison Excel against a deck SUPPRESS with "Ignore deck-pipeline".
As a process C 62/100 · Has gaps — weak spots: result and completion, inputs and preconditions
How to improve
- 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 · 0
✓ No critical or high findings
Files scanned: 16. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 62/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 70Execution cost. Instruction body is 4072 tokens
- 100Steps. 85 steps
- 100Failures and branches. 3 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 The response is described with custom markup (22 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)
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +5Description quotes 4 example trigger phrases
- +3Description length 720: enough signal without eating the budget
- +4Structure: 26 headings
- +3Step-by-step instructions: 85 items
- +4Has examples (1 code blocks)
- +3All 10 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 93.