AB polling-best-practices
Best practices for automating long-running, asynchronous tasks via cron-style polling. Use when the user wants to monitor a background CLI command or API call that does not return an immediate result — such as waiting for a report to generate, a podcast to be created, a deep research job to complete, or any task with a clear completion condition. Triggers on requests like "monitor X until done", "poll for Y", "wait for Z to complete in the background", "run X in the background and notify me when ready", "long-running task automation", "async task monitor". This skill is NOT for tasks requiring real-time interaction, complex LLM judgment during execution, or streaming output. It is a guidance-only skill — it does not install scripts or tools; the agent applies these principles manually based on the guidance here. A template polling script is provided in the scripts/ directory for convenience.
As a process B 66/100 · Nearly there — weak spots: result and completion, inputs and preconditions
The same skill appears in 1 more place: ClawHub
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
-
low Dangerous commands
cmd-background-processSKILL.md:219Starts a background / autostarted processnohup bash poll.sh > /dev/null 2>&1 &
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 66/100
- 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
- 85Steps. 43 steps, 1 vague phrases
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 9 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2637 tokens
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
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
- +3Description length 904: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 6 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 16 headings
- +3Step-by-step instructions: 43 items
- +4Has examples (8 code blocks)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 93.