InfernoFirewall is an inline gateway between agents and LLMs. The ML layer proposes facts; Cedar policy decides the verdict. A tenant with no active bundle has no access, so the default posture is deny-by-default. This manual covers what operators need to run, audit, and trust the system.
fw migrate as the database owner, then seed a tenant, provider route, model catalog entry, and gateway API key.fw bundle build validates, checks invariants, diffs against the active bundle, and stores the candidate bundle.fw audit verify checks hash chains, checkpoints, gaps, edits, forged records, and pinned public keys.Policies are Cedar rules stored per tenant. The baseline guard rails are included in every bundle: secrets do not reach external sinks, unreviewed tools cannot run, irreversible actions need approval, and high injection scores block tool calls.
A tool call is allowed only when a matching permit exists and no forbid rule blocks it. If evaluation errors, the result is deny.
Classifier facts appear in forbid conditions. The policy ceiling can shrink, but it cannot widen.
Allow, deny, step-up, redact, sandbox, and fail-closed outcomes each carry reason codes, rules fired, scores, thresholds, and stage latency.
An approval is bound to tenant, session, principal, tool pin, and canonical arguments hash. A changed amount is a different call.
A session is where provenance matters. Anything returned by a tool is untrusted for the rest of that session, and policy decides what untrusted content means for later calls.
Reads, fetches, tool results, and child agents inherit taint and data classes. Declassification is an audited, append-only action.
MCP tool lists are approved only when the tool description, schema, and annotations match the recorded pin. If they change upstream, the tool disappears until re-review.
The result is scanned and recorded first. If the session state cannot be saved, the result is withheld.
A consequential allowed or redacted call waits for its audit record. If the commit fails or times out, the call is refused before any approval is spent.
The audit log is not a side channel. It is part of enforcement: consequential actions are recorded before they run, and every decision can be replayed against the bundle that produced it.
fw audit keygen fw audit verify --tenant demo --public-key <pinned-public-key> fw replay --tenant demo --limit 1000 fw replay --tenant demo --bundle <candidate-bundle-hash>
Replay compares stored Cedar inputs and recorded verdicts. Detection scores can be stored without raw text, so detection replay depends on payload retention and is not assumed.
Records are chained by SHA-256 over the previous hash and canonical record, one chain per tenant and gateway node.
Ed25519 checkpoints anchor the chain. Verification can be pinned to a public key kept outside the database.
Export sinks move cursors only after acknowledgement, so outage causes lag but not skipped records.
Self-hosted today: one Rust binary and Postgres. Docker Compose is available for quick start, while production upgrades should back up first, run migrations, roll gateways, then verify the chain.
| Area | Default posture |
|---|---|
| Decision pipeline | Hard deadline; tool calls fail closed on timeout, stage error, panic, malformed JSON, or lost session state. |
| Plain text | Can fail open only if FW_FAIL_OPEN_TEXT is explicitly enabled. |
| Private provider addresses | Normal model traffic follows FW_UPSTREAM_DENY_PRIVATE; console probes use FW_PROBE_DENY_PRIVATE. |
| Console access | Role-gated. Payload views write access records before content is shown. |
| Provider keys | Stored encrypted under FW_MASTER_KEY, bound to tenant, provider, and base URL. |