SKILLEMALL.ai

AA fde-deployment-architect

Stage 4 of FDE Delivery Loop. Turn an approved POC PRD into a runnable, risk-controlled deployment architecture and delivery plan covering system boundaries, data, permissions, integrations, environments, observability, and risk. Use for POC implementation planning, architecture review, data and permission mapping, integration design, and pre-deployment risk decisions. Do not use to rewrite business requirements, design agent behavior, or replace a production architecture review.

ClawHub Agent Skills author: xukun0821 v1.0.0 MIT-0 15 files body ≈ 1 557 tokens Open the sourceclawhub.ai analyzed 35 h ago

Stage 4 of FDE Delivery Loop.

As a process A 81/100 · Runs to the end — weak spots: running it twice, progress reporting

IntegrationInfrastructuretype and topics are labelled automatically from the skill text
JSON
Technical rating
A
97/100
safety, quality, tests
Safety 60%
100
Quality 40%
92
Run on models
none yet
Process rating
A
81/100
Runs to the end
Progress reporting w 2
0
Running it twice w 4
30
When it triggers w 12
50
the three weakest of ten parameters · all ten

How to improve

    For the model run — optional
    • 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: 15. 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 81/100

    • 0Progress reporting. Says nothing while it works
    • 30Running it twice. 8 mutating operations with no state check
    • 50When it triggers. No condition that starts the skill
    • 50Failures and branches. 0 branches, has a failure section
    • 70Inputs and preconditions. Inputs and preconditions are listed
    • 100Tools and files. No external tools needed
    • 100Steps. 28 steps
    • 100Result and completion. Output format and completion criterion are stated
    • 100Consistency. Name and required fields are in place
    • 100Execution cost. Instruction body is 1557 tokens

    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
    • +4No input/output examples
    • +2Single-language instructions
    • +4Description says when NOT to use the skill
    • +3Description length 484: enough signal without eating the budget
    • +4Structure: 9 headings
    • +3Step-by-step instructions: 28 items
    • +3Output format is stated explicitly
    • +4Reference files are cited in the instructions (10 of 10)
    • +1License stated

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 92.

    External checks

    ClawHub: clean
    This skill is a planning guide for POC deployment architecture and risk review, with no executable install steps or hidden system-changing behavior.
    LLM: benign (high) · VirusTotal: · 7 Aug 2026