AB prompt-archeologist
Reverse-engineers high-quality, reusable prompts from messy conversations, vague requests, or rough user descriptions. Use this skill whenever a user: wants to "save" or "capture" what they've been doing in a conversation as a reusable prompt; says things like "how do I ask this again?", "turn this into a prompt", "what prompt should I use for this?", "extract the prompt from this conversation", or "I want to recreate this later"; shares a messy or rambling description of a task and wants it cleaned up into something repeatable; or asks for help building a prompt library, template, or reusable instruction set. Always trigger on any variation of "make this a prompt", "save this workflow", or "what did I just do?" — even if the word "prompt" is never used.
As a process B 76/100 · Nearly there — weak spots: inputs and preconditions, progress reporting
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: 2. 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 76/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 85Steps. 9 steps, 2 vague phrases
- 100Tools and files. No external tools needed
- 100Failures and branches. 8 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1632 tokens
- 100Running it twice. No mutating operations
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)
- +1No license
- +2Single-language instructions
- +5Description quotes 10 example trigger phrases
- +3Description length 764: enough signal without eating the budget
- +4Structure: 12 headings
- +3Step-by-step instructions: 9 items
- +3Output format is stated explicitly
- +4Has examples (4 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 92.