vise --help or vise <command> --help. Most agent-facing commands have an MCP tool equivalent in snake_case; trust signing and Managed Placement installation are intentionally CLI-only.
Most commands take a project path (default: the current directory) and read or write state under
sp-vise/. The ones reviewers reach for most: vise impact, vise status, vise check --ci, and vise passport.Explore and plan
| Command | Purpose |
|---|---|
vise explore "<request>" | Map a request to what social.plus offers — no project or API key needed |
vise inspect [path] | Detect platform, monorepo surfaces, design signals, available sensors, and bounded scan coverage. An incomplete scan cannot initialize or produce a green check; select the concrete app surface in a large monorepo |
vise plan [path] --request "..." | Grounded implementation plan with intake questions and docs citations (--summary for a compact view) |
vise init [path] --request "..." --answer key=value | Write the sp-vise/ compliance contract once blocking intake is answered. Reviewed organization policy, signed-policy trust, and aggregate measurement intent can be accepted with --policy-pack, --trust-policy, and --measurement-plan. Initializing a different outcome records the previous engagement to sp-vise/engagements/ first; replacing a green engagement asks for --supersede-engagement |
| Command | Purpose |
|---|---|
vise workplan next [path] --request "..." | Print the next uncompleted surface and its focused commands |
vise workplan status [path] --request "..." | Show the sequence and which surfaces are done |
vise workplan complete [path] --request "..." --surface <id> | Record a green-checked surface with snapshot evidence |
vise workplan trim [path] --request "..." --surface <id> --reason "..." | Deliberately omit a companion surface and re-arm blueprint confirmation so the changed journey is reviewed |
Day-2 and release assurance
| Command | Purpose |
|---|---|
vise status [path] | Read-only lifecycle front door for build, Day-2 change, and release review. Keeps compliance, runtime, SDK, policy, advisory, Passport, and human-approval claims separate; reports owners and the smallest next actions without running sensors, writing files, or changing the compliance exit code |
vise impact [path] [--symptom "..."] | Read-only Day-2 planner. Compare the project with recorded intent, lifecycle state, SDK declarations, policy, and evidence; classify evidence as preserved, stale, missing, or unknown; and return the smallest safe recovery sequence. A symptom can authorize a bounded repair when recorded cross-surface intent and current source confirm the gap |
vise sdk-release [path] | Explain whether detected SDK declarations and supported lockfile resolutions match Vise’s exact extraction-grounded snapshots, reviewed knowledge, and five-platform fact-depth contract. meets-target, target-gap, regressed, and unknown stay explicit; it does not fetch registries, edit dependencies, or infer semantic compatibility from semver |
vise sdk-release --from-surface-dir <before> --to-surface-dir <after> | Diff two SDK surface snapshots for symbol, model-field, capability, and affected-rule changes |
vise policy status [path] [--source <manifest>] [--ci] | Read-only organization-policy source, lifecycle, trust-key rotation/revocation, deterministic update impact, and exact owner routing. CI exits 0, 12, or 13 for ready, review-required, or blocked |
vise policy accept [path] --source <manifest> --recorded-by <identity> --reason "..." | CLI-only explicit write: pin the exact reviewed source and Policy Pack digests and write an audit receipt. It never rewrites policy inputs, manages keys, or asserts release approval |
vise passport [path] [--write] | Compose sanitized accepted intent, lifecycle, SDK assurance, evidence hashes, runtime state, residual risk, and replay commands into digest-bound claims. Read-only unless --write is explicit |
vise passport verify [path] [--passport <path>] | Recompute Passport claims and local evidence states without rerunning or repairing the project |
vise passport compare [path] --from <before> --to <after> | Compare two valid Passports and compose a Governed Release Review result: ready, review-required, or blocked |
vise trust sign [path] ... | CLI-only explicit write: wrap a reviewed Passport or Policy Pack in an Ed25519 envelope using an external private key that Vise does not retain |
vise trust verify [path] --envelope <path> --trust-policy <path> | Authenticate envelope integrity, issuer/key authorization, artifact type, audience, and validity. Signature validity does not replace Passport or Policy Pack semantic validation |
Validate and verify
| Command | Purpose |
|---|---|
vise check [path] | Validate the code against the recorded contract |
vise check [path] --ci | Read-only release gate for the active contract and every recorded engagement; exits 11 when active work is green but completed work drifted |
vise check [path] --ci --from-passport <before> --to-passport <after> | Add Governed Release Review to an otherwise unchanged compliance gate. Existing compliance codes take precedence; otherwise-green compliance exits 12 when release evidence is review-required or blocked. Add --trust-policy when both inputs are signed envelopes |
vise check [path] --new-only | Gate only on rule findings introduced since the baseline — the outcome’s completeness checklist and selected capabilities always gate (they are the engagement’s deliverables) |
vise check [path] --allow-proof-waiver | Accept an honest runtime-proof-waived result as passing |
vise check [path] --engagement <outcome> | Re-verify one previously completed engagement against the current rules and code (exit 11 on drift, 0 when it still holds) |
vise check [path] --all-engagements | The normal check plus re-verification of every recorded engagement; exit 11 means your current work is green but a previously finished surface drifted |
vise validate [path] | Run the deterministic validators only (no attestation comparison) |
vise explain <ruleId> | Print a rule’s rationale, evidence requirements, and remediation |
Evidence
| Command | Purpose |
|---|---|
vise baseline [path] | Snapshot pre-existing rule findings so --new-only can exclude them. Record it right after vise init, before writing code; the feature you asked for is never baselined — it still has to be built |
vise attest [path] --rule <id> --signer host-agent --confidence high --evidence-file evidence.json --rationale "..." | Record that a rule is satisfied through architecture the check can’t see |
vise sync [path] | Persist deterministic-pass evidence after a green check |
vise smoke declare [path] --surface <id> [--route <route>] [--target-type <type>] | Declare the exact runtime collection surface and, when applicable, its route and SDK target |
vise smoke [path] --log <file> [--surface <id> ...] [--receipt-context <json>] | Assess normalized runtime markers and write independent per-surface receipts. Marker surface, route, and target must match the declaration; strict receipt context additionally binds source, build, runner, artifacts, and prior Passport lineage |
vise smoke verify [path] --evidence <scoped-runtime-smoke.json> | Read-only receipt verification without relaunching the app; distinguishes absent, stale, tampered, failed, and passed |
vise smoke waive [path] --reason "..." --mode declined|deviceless | Record an auditable runtime-proof waiver |
Engagement history
| Command | Purpose |
|---|---|
vise engagement init [path] --tier <tier> --customer-id <id> --scope <outcome,...> | Record the contractual broad-social engagement scope, with optional target date, reviewer, and evidence-upload consent |
vise engagement show [path] | Read the recorded engagement-scope artifact |
Project sensors
| Command | Purpose |
|---|---|
vise run-sensors [path] [--dry-run] [--include "<sensor name>"] | Run detected project build, lint, typecheck, test, and SDK-smoke commands. An explicit --include must match a detected sensor or the command fails; captured output is bounded, and timed-out sensors terminate their descendant processes |
Docs and SDK facts
| Command | Purpose |
|---|---|
vise search-docs "<query>" | Search social.plus docs for relevant pages |
vise get-doc-page <path> | Fetch a specific doc page by path |
vise sdk-facts --platform <platform> | Read bundled SDK surface facts (--capability, --format json) |
vise debug [path] --error "..." | Diagnose an SDK-specific runtime failure |
vise placements plan [path] --placement <id> --package-source <local-path> | Plan explicitly selected Managed Placement host and slot governance without writing |
vise placements add [path] --placement <id> --package-source <local-path> [--dry-run|--apply] | Preview by default, or explicitly install the pre-launch private host from a local/file source. Vise never guesses a slot coordinate |
vise placements validate [path] [--placement <id>] | Validate one host initialization, explicit slot/context evidence, secret hygiene, and sidecar drift |
Experience planning and outcome evidence (advisory)
| Command | Purpose |
|---|---|
vise creative [path] --request "..." | Produce an Engagement Intelligence brief with candidate experiences, rationale, tradeoffs, and review gaps |
vise creative accept [path] --variant <id|none> ... | Record an explicit, reasoned selection or catalog gap |
vise ux-harness [path] | Generate advisory UX expectations from the selected variant |
vise experience compile [path] | Compile the selected variant into implementation artifacts |
vise experience sensors [path] | Write the advisory experience sensor framework |
vise experience-report [path] | Write a dimensioned advisory experience review |
vise learning record [path] | Record a local-only learning event or reviewed aggregate Outcome Evidence |
vise learning record [path] --kind incident-protection --evidence <input.json> [--no-write] | Preview or record a reviewed, sanitized incident bound to distinct finding/clean fixtures and exact evidence. It never executes replay, uploads source, edits protection, or changes readiness |
vise learning show [path] | Read the local learning summary and non-causal advisory deltas |
vise check exit codes.
Design (advisory)
| Command | Purpose |
|---|---|
vise design extract <prototype> | Extract a graded design contract from an HTML/CSS prototype (--from-project to derive from your own tokens) |
vise design check [path] | Advisory token-conformance report, including computed WCAG for the brand palette |
vise design preview [path] | Write a self-contained visual review for human or VLM judgment |
Skill and MCP
| Command | Purpose |
|---|---|
vise install-skill --target <host> | Install the bundled skill (claude, cursor, vscode, copilot, codex) |
vise print-skill | Print the skill markdown to stdout |
vise mcp | Start the local MCP server |
vise doctor | Verify the install — version, path, docs source |
Related
How Vise works
The loop behind these commands and how to read results.
Run Vise as an MCP server
The MCP tool equivalents of these commands.