Repository navigation
Demos Pocketos
After a public incident report: in 2026 a coding agent that hit a credential mismatch deleted PocketOS's
production database with one call to Railway's GraphQL API (volumeDelete). Here a scripted coding agent makes that
call as a Bash tool call under the dev pack, the signer's default for coding agents on a laptop. The curl waits for
a person (TK-A001, destroying a deployment, database or volume). The person rejects it with a reason; the agent's
railway and git calls run.
python examples/v2/demos/pocketos.py # POSIX
tracekit verify run.tkb --trust trust.jsonWhat the record shows. The curl's policy.decision (ask and the rule ids) before it could run, the person's
approval (reject, with the reason they gave), then the signer's approval.refused (TK-APPROVAL-REJECTED) when the
agent tries to run it anyway; no tool.result for it. A call the person approves would show the approval, then its
result.
Why it matters. The rule holds the call itself, whatever the model says it is doing: a volumeDelete( mutation
on the command line of curl, wget, HTTPie, xh or railway api waits for a person. It reads the command as
typed, so the same mutation in a body read from a file (-d @mutation.json) or sent from an interpreter one-liner
isn't held (what TK-A001 reads); give an agent no token that can delete
production volumes.
transcript.txt · the example · packs · limits
Generated from docs/ at v1.4.2. Edit the docs in the repository, not here.
Get started
- Quickstart: the v2 signer
- Quickstarts
- Quickstart: Claude Agent SDK
- Quickstart: a custom agent
- Quickstart: LangChain and LangGraph
- Quickstart: MCP client
- Quickstart: OpenAI Agents SDK
- Coding agents other than Claude Code
- tracekit doctor
- Walkthrough: Tracekit end to end
How it works
- How Tracekit works
- Overview
- The life of a tool call
- The evidence format
- Verification and assurance
- Witnesses and monitoring
- Policy
- Approvals
- Detection and containment
- Capture and completeness
- Custody
- Provenance and tracekit why
- Deployment and isolation
- Storage, durability and performance
- Integrations
- Security model
- Architecture
- tracekit why: architecture
Integrations
- Integrations
- Adapters
- OpenTelemetry
- Remote clients
- Kubernetes agent-sandbox
- agentgateway
- agentsh and Tracekit
- Browser Use
- Claude Agent SDK
- Claude Code's sandbox
- Codex's sandbox
- CrewAI
- Docker MCP Gateway
- E2B
- Falco
- Gemini CLI's sandbox
- Google ADK
- Haystack
- Hermes Agent
- LangChain v1
- LiteLLM
- LlamaIndex
- MCP client
- microsandbox
- Microsoft Agent Framework
- OpenAI Agents SDK
- NVIDIA OpenShell
- Pydantic AI
- sandbox-runtime's proxy on its own
- Strands Agents
- Tetragon
- Vercel AI SDK
- Wassette
Policy and approvals
- Policy reference
- Approvals
- Findings
- Reading the model's reasoning
- Incidents and containment
- Webhooks
Deploy and operate
- Deploying the signer
- The signer as a container
- The signer on Kubernetes
- The compose stack
- Identity
- Tool executor
- MCP gateway
- LLM gateway
- Egress observation
- Witnesses
- Log monitor
- Record-key issuer
- Signer observability
- Viewer
- tracekit why: causal investigation and replay
- Migrating
- Platforms
Verify the evidence
- Tracekit for auditors
- Reading a v2 verify report
- Evidence format v2
- Auditor guide
- Sample evidence bundle
- Signing
- What Tracekit guarantees
Security and privacy
- Threat model: the v2 signer
- Threat model: the v1 laptop signer
- HTTP security checklist
- Threat-model review packet
- Privacy
- Known limits and trade-offs
- What's stable
Compliance
- Compliance mapping
- EU AI Act
- ISO/IEC 42001
- NIST AI RMF and the Generative AI Profile
- OWASP Top 10 lists for LLM and agentic applications
Evaluation
Demos
- Demos
- Agent approvers within bounds
- A production bucket
- Completeness and tamper evidence
- Detection and containment
- Network coverage: an egress observer
- Guardrails: allow, deny, or ask a person
- MCP tools, gated and recorded
- A production volume
- Prompt injection held by provenance
- Why: which input caused the action
Help