AC sandbox-selfheal-guard
Keep local GGUF/llama.cpp agent sandboxes from hanging after snapshot eviction. Use when: inference hangs or "thinks forever", binaries/models vanished from ~/llama.cpp or failed byte-checks, npx stalls on closed stdin, or sudo might prompt interactively.
As a process C 62/100 · Has gaps — weak spots: result and completion, inputs and preconditions
How to improve
For the model run — optional
- 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: 11. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 5, column 14: description: Keep local GGUF/llama.cpp agent sandboxes from hanging after snaps… ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value
Process rating: all ten parameters 62/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 100Steps. 16 steps
- 100Failures and branches. 3 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1456 tokens
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +4No input/output examples
- +2Single-language instructions
- +3Description length 255: enough signal without eating the budget
- +4Structure: 8 headings
- +3Step-by-step instructions: 16 items
- +3All 5 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 74.
External checks
ClawHub: suspicious
The skill has a coherent self-repair purpose, but its fix mode gives broad system, network, build, and file-write authority that is not tightly enforced.
LLM: suspicious (high)