faramesh.devBETA
docsgithubcommunityget started →

docs

Getting Started

Core

FPL

Governance

Operations

Reference

github ↗slack ↗community ↗

13 , incidents

Incident Response

Faramesh provides a structured incident workflow: declare an incident, isolate the agent, collect evidence, and resolve. All incident actions are recorded in the DPR chain for audit.

Incident workflow

The four-phase incident lifecycle.

DeclareCreate an incident record with severity and description
IsolateKill-switch the agent and quarantine its sessions
EvidenceAttach audit exports, logs, and context to the incident
ResolveClose the incident with a resolution summary

Declare an incident

Create a new incident with severity and description.

bash
faramesh incident declare \
  --severity high \
  --agent support-bot \
  --reason "Agent issuing refunds without customer request"

Isolate the agent

Immediately deny all actions from the agent and quarantine its sessions.

bash
faramesh incident isolate <agent-id>

Attach evidence

Attach audit exports, screenshots, or other evidence to the incident record.

bash
faramesh incident evidence <incident-id>

Resolve the incident

Close the incident with a resolution summary. The agent remains isolated until explicitly unkilled.

bash
faramesh incident resolve <incident-id>

Run a playbook

Playbooks automate common incident response workflows.

bash
faramesh incident playbook <incident-id>

All incident actions (declare, isolate, evidence, resolve) are recorded in the DPR chain. The incident ID links all related audit records together.

Need help? Use any of these support options.

open an issuejoin slackcommunity/forum
MIT License
docsgithubcommunityget started →