Readonly diagnostics runner
Local-first Agent Skill Runtime for packaging SOPs, schemas, preflight checks, run records, and MCP tools as executable skill capsules.
npx -y skills add iiwish/skillrun --skill readonly_diagnostics_runnerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.1 KB, 243 tokens by cl100k_base, as published. Nobody here has run it
Read-only Diagnostics Runner
Purpose
This SkillRun capsule runs a small allowlist of read-only diagnostic commands. It demonstrates how a high-risk terminal-shaped action can be narrowed into a typed skill contract.
SOP
- Accept only named diagnostics from the schema. Do not accept arbitrary shell strings.
- Run commands without
shell=True. - Do not support redirection, pipes, command chaining, mutation commands, package installs or process killing.
- Capture stdout, stderr, exit code and timeout status.
- Truncate oversized output so the Agent can recover without flooding context.
Allowed Diagnostics
pwd: print the capsule working directory.list: list files in the capsule working directory.git_status: rungit status --shortwhen the current directory is inside a Git repository.
Recovery Guidance
If a diagnostic fails, inspect the structured exit_code, stdout and
stderr. Do not retry with a broader shell command.
Prohibited Behavior
- Do not pass arbitrary command strings.
- Do not run destructive commands.
- Do not treat this capsule as a general-purpose shell or an OS sandbox.
What ships with it: 4 files
3.3 KB alongside SKILL.md, 1 of them executable
examples/
- action.pyruns2.9 KB
- skillrun.config.json286 B