AC travel-lobster
Autonomous internet exploration skill. Your agent roams the web driven by its own curiosity, discovers interesting things, and sends illustrated "postcards" — personal letters with AI-generated art — to a chat. Features persistent travel memory with knowledge graph, curiosity seeds, growth tracking, time-aware tone, and self-scheduling random-interval trips. Inspired by "Travel Frog" (旅行青蛙). Activate when user asks to explore the internet autonomously, send postcards, discover interesting things, or be a "travel frog/lobster".
As a process C 63/100 · Has gaps — weak spots: result and completion, when it triggers, running it twice
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 skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
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 · 2
✓ No critical or high findings
Medium and low: 2
-
medium Dangerous commands
cmd-persistenceSKILL.md:236Persistence mechanism (cron / launchd / scheduled task / autorun registry) (detector / deny-list definition)crontab -l | grep -v watchdog | crontab -
detector -
low Dangerous commands
cmd-cron-mentionSKILL.md:236Mentions editing / listing crontabcrontab -l | grep -v watchdog | crontab -
Files scanned: 7. 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
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 7 mutating operations with no state check
- 60Tools and files. Uses tools (bash, python) that frontmatter does not declare
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 39 steps
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2844 tokens
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 11 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (4 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
- -215 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +3Description length 532: enough signal without eating the budget
- +4Structure: 27 headings
- +3Step-by-step instructions: 39 items
- +4Has examples (6 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
- +3All 4 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 94.