AC puzle-read-skill
Connect to Puzle Read — an intelligent reading workbench that helps users turn articles and documents into searchable personal knowledge. Save web articles (by URL or pre-fetched content), upload documents and files to the user's reading library, and search across everything they've read to find relevant insights. Use this skill whenever the user wants to: save or read web articles, add documents to their reading library, look up something from their reading collection, process PDFs or other files for deeper understanding, or build knowledge workflows around their readings. This skill also applies when the user wants to: save something for later reading ("read it later", "bookmark this", "save this article"), get a summary or analysis of an article or document, upload a file (PDF, image, audio, etc.) for processing, or organize their reading collection. Also trigger when you see imports of `puzle_reading` or `PuzleReadingClient`, or when the user mentions Puzle.
Connect to Puzle Read — an intelligent reading workbench that helps users turn articles and documents into searchable personal knowledge.
As a process C 63/100 · Has gaps — weak spots: result and completion, consistency
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
- 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 · 1
✓ No critical or high findings
Medium and low: 1
-
medium Exfiltration
net-redirectable-api-keyscripts/puzle_reading.py:79Helper 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
Files scanned: 3. 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 63/100
- 0Result and completion. Does not say what the result is
- 40Consistency. Frontmatter name (puzle-read-skill) differs from the folder (puzle-read)
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 4559 tokens
- 85Steps. 39 steps, 1 vague phrases
- 100Failures and branches. 3 branches, has a failure section
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- low 10 top-level sections: this looks like several domains in one skill
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 976: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +4Structure: 37 headings
- +3Step-by-step instructions: 39 items
- +4Has examples (23 code blocks)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 89.