Update last activity timestamp.

Record activity for an agent (resets idle timer)

Record API request.

Record an audit log entry.

Record a chaos experiment as an OTel span with fault details.

Record collaboration health signal.

Record a cost observation for a trace. In live mode, creates a Langfuse span with cost metadata. In offline mode, stores the observat

Record credential issuance.

Record credential revocation.

Record that data was accessed (for audit).

Observe a scope chain depth. Args: depth: Depth of the scope chain.

Record a scope chain depth observation. Args: depth: Depth of the scope chain.

Record a dispute outcome and update history.

Record a trust handshake.

Record a policy evaluation. Args: policy_name: Identifier of the evaluated policy. action: Decision outcome (e.g. ``"

Advance the SnapshotBuilder budgets after a successful call. Mirrors the v4 ``ctx.call_count += 1`` / ``ctx.total_tokens += usage`` p

Record a fault impact score from a chaos experiment. Args: experiment_name: Name of the chaos experiment score: Fault

Record resource efficiency signal.

Record a reward signal.

Record security posture signal.

Record metrics directly from an SLO object. Args: slo: An agent_sre.slo.objectives.SLO instance

Record successful interaction, increasing trust.

Record ticket handling metrics.

Record a tool call.

Register an extractor for sub-objects stored at `contextKey`.

Register an agent with trust verification.

Register an agent blueprint for consistent identity creation.

Register a capability that this agent can perform. Args: capability: Name of the capability handler:

Register a regex pattern that maps matching tool names. Args: pattern: Regex pattern to match tool names against. can

Register an external peer's identity (no private key).

Register an SLI type for discovery.

Reject a review item. Args: review_id: ID of the review to reject reviewer_notes: Required notes exp

Release escrow based on outcome. - success: Credits go to provider, provider reputation +2 - failure: Credits return

Release a distributed lock

Remove an agent from the directory. Args: did: The agent's DID Returns: True if removed, False if not found

Remove an agent from inventory.

Remove a policy by name. Returns true if found and removed.

Remove an agent DID from a namespace. Args: namespace_name: Name of the target namespace. agent_did: DID of the agent

Remove an org policy. Returns True if found.

Remove a policy from the engine. Args: name: Name of the policy to remove. Returns: ``True`` if the policy w

Remove a principle by name. Returns True if found and removed.

Remove a role assignment, reverting the agent to the default role.

Remove a session's isolation scope (session ended).

Remove an SLO by name.

* Fallback Markdown-to-HTML renderer for older VS Code hosts where * `markdown.api.render` is unavailable. Supports headers, lists, tables, * inline

Render a polymorphic response as React JSX (pseudo-code). Args: response: The polymorphic response from the agen

Replay persisted messages from a topic (AMB-001). Args: topic: Topic to replay messages from handler

Replay the last N minutes of an agent's life. Args: agent_id: Agent identifier minutes: Number of mi

Send a JSON-RPC request and wait for the matching response.

Reset bootstrap state (for testing only).

Reset the bucket for a specific agent/action pair, or all buckets.

Reset limits for an agent or all agents. Args: agent_did: If provided, reset only that agent. Otherwise reset everything.

Reset all attributes to their initial values.

Reset baselines and windows for *metric_name*, or all if ``None``.

Reset the global profiling report.

Wait for and return the human decision on an escalation. Delegates to the handler's resolve method, which blocks or polls depending o

Resolve injection markers in function parameters. Args: func: The function being called. args: Positional arguments p

Process a chat message and return a result dict.

Restore VFS to a previous snapshot. Args: snapshot_id: ID returned by create_vfs_snapshot. agent_did: Agent requestin

Retrieve episodes, optionally filtered.

Governed retrieve.

Retry processing a failed message. Args: message_id: Message ID to retry handler: Handler function t

Review patient chart with full HIPAA compliance.

Remove all endorsements from a specific endorser for a target. Returns: Number of endorsements removed.

Revoke this grant immediately.

Revoke this agreement.

Revoke all active grants in this scope. Returns: Number of grants that were revoked.

Revoke all credentials for an agent. Args: agent_did: The agent's DID whose credentials should be revoked. reason: Hu

Revoke all grants issued by a specific grantor. Useful when a grantor agent is compromised and all grants it issued must be invalidat