BC solo-impl
Autonomous executor for Solo CLI — actually runs setup, calibration, teleoperation, dataset recording, policy training, and inference commands in the user's terminal using Shell tool. Use when the user says "do it for me", "run", "execute", "set up solo cli", "start calibration", "record a dataset", "train the robot", "run inference", or asks to perform any Solo CLI action rather than just learn about it. Works standalone — does not require solo_cli_guide to be installed.
As a process C 56/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
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
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- 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 · 8
-
high Dangerous commands
cmd-pipe-to-shellSKILL.md:159Downloads and executes remote code from an unrecognised host (pipe to shell)- **curl | sh disclosure** — Before running the uv installer the agent shows the exact URL (`https://astral.sh/uv/install.sh`) and explains it is the official uv installer, then runs it. The user can
Medium and low: 7
-
medium Dangerous commands
cmd-pipe-to-shellprompts/impl_executor_prompt.txt:24Downloads and executes remote code from an unrecognised host (pipe to shell) (detector / deny-list definition)### Rule T1 — curl | sh Explicit Consent
detector -
medium Dangerous commands
cmd-pipe-to-shellskill.json:43Downloads and executes remote code from an unrecognised host (pipe to shell) (string literal in code, not executed)"method": "curl | sh",
code literal -
low Dangerous commands
cmd-pipe-to-shell-known-hostdomains/environment.json:7Pipe-to-shell installer from a well-known host (still executes remote code) (string literal in code, not executed)"macos": "curl -LsSf https://astral.sh/uv/install.sh | sh",
code literal -
low Dangerous commands
cmd-pipe-to-shell-known-hostdomains/environment.json:8Pipe-to-shell installer from a well-known host (still executes remote code) (string literal in code, not executed)"linux": "curl -LsSf https://astral.sh/uv/install.sh | sh",
code literal -
low Dangerous commands
cmd-execpolicy-bypassdomains/environment.json:9Runs PowerShell with execution policy bypassed (quoted — discussed, not commanded)"windows": "powershell -ExecutionPolicy ByPass -c \"irm https://astral.sh/uv/install.ps1 | iex\""
quoted -
low Dangerous commands
cmd-pipe-to-shell-known-hostprompts/impl_executor_prompt.txt:31Pipe-to-shell installer from a well-known host (still executes remote code) (quoted — discussed, not commanded)> `curl -LsSf https://astral.sh/uv/install.sh | sh`
quoted -
low Dangerous commands
cmd-pipe-to-shell-known-hostSKILL.md:176Pipe-to-shell installer from a well-known host (still executes remote code) (documentation table row)| `curl -LsSf https://astral.sh/uv/install.sh \| sh` | astral.sh | Official uv installer | Yes — URL and purpose shown to user |
table
Files scanned: 11. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 56/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 10 mutating operations with no state check
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, web, git, python) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 100Steps. 34 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2216 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 12 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)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 7 example trigger phrases
- +3Description length 476: enough signal without eating the budget
- +4Structure: 16 headings
- +3Step-by-step instructions: 34 items
- +4Has examples (3 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 88.