vulkro assess-change
Assess whether a change is safe and non-breaking, for an AI agent.
Command reference
Generated from vulkro assess-change --help (vulkro 0.36.0). Run scripts/gen-docs.sh in the product repository to refresh it; do not edit this section by hand.
Assess whether a change is safe and non-breaking, for an AI agent.
vulkro assess-change [OPTIONS] [PATH]
Arguments
| Argument | Description |
|---|---|
[PATH] | Path to the project root (default: current directory) Default: .. |
Options
| Flag | Description |
|---|---|
--offline | Hard-disable every outbound network call for this run (sets VULKRO_OFFLINE=1). Blocks the CVE feed, the license heartbeat, the update check, webhooks, and any cloud AI endpoint; a loopback model (http://127.0.0.1, http://localhost) is still allowed. Equivalent to exporting VULKRO_OFFLINE=1, and the flag wins when both are set. |
--symbol <SYMBOL> | Pre-change mode: the symbol about to be edited (file::name or bare name). Omit for post-change (diff) mode. |
--base <BASE> | Post-change mode: a git ref to diff against (base...HEAD). Omit to assess the uncommitted working-tree change (git diff HEAD). Use e.g. HEAD~1 for the last commit or origin/main for a branch. |
-q, --quiet | Print only warnings and errors on stderr: no progress, tips or next-step hints. Results on stdout and exit codes are unchanged. |
--no-color | Turn off coloured output. Colour is also off when NO_COLOR is set to a non-empty value or TERM=dumb. |
-h, --help | Print help (see a summary with '-h') |
Full vulkro assess-change --help
Assess whether a change is safe and non-breaking, for an AI agent.
Two modes. Pre-change: pass `--symbol NAME` to see, before you edit it, its blast radius (what transitively depends on it), which tests to run, and whether it already carries findings or sits on a taint path. Post-change: pass no symbol to assess the current diff (`--scope`), which reports the dependents and tests the change touches and the NEW findings it introduced, with a "safe to proceed / review N" verdict.
Emits deterministic JSON. Structural impact is a floor; the security half reflects a full scan of the tree.
Exit code: 0 assessment printed, 2 IO / argument / git error.
Usage: vulkro assess-change [OPTIONS] [PATH]
Arguments:
[PATH]
Path to the project root (default: current directory)
[default: .]
Options:
--offline
Hard-disable every outbound network call for this run (sets VULKRO_OFFLINE=1). Blocks the CVE feed, the license heartbeat, the update check, webhooks, and any cloud AI endpoint; a loopback model (http://127.0.0.1, http://localhost) is still allowed. Equivalent to exporting VULKRO_OFFLINE=1, and the flag wins when both are set
--symbol <SYMBOL>
Pre-change mode: the symbol about to be edited (`file::name` or bare `name`). Omit for post-change (diff) mode
--base <BASE>
Post-change mode: a git ref to diff against (base...HEAD). Omit to assess the uncommitted working-tree change (git diff HEAD). Use e.g. `HEAD~1` for the last commit or `origin/main` for a branch
-q, --quiet
Print only warnings and errors on stderr: no progress, tips or next-step hints. Results on stdout and exit codes are unchanged
--no-color
Turn off coloured output. Colour is also off when NO_COLOR is set to a non-empty value or TERM=dumb
-h, --help
Print help (see a summary with '-h')