docs/archive/.
Launch Objective
Ship a credible beta for teams that need runtime control of AI agents:- instrument Python agents quickly;
- enforce budgets, policies, guardrails, approvals, and kill switches;
- sync telemetry to the platform;
- operate from the console;
- prove what happened through the ledger, exports, replay, and compliance metadata.
Positioning
Agent Sentinel is runtime authority for AI agents. Keep product language focused on shipped capabilities:- Use: runtime policy enforcement, intervention tracking, approvals, replay, ledger, cost controls, kill switch, and compliance/audit workflows.
- Avoid: unqualified certification claims such as “SOC 2 certified” unless certification exists.
- Clarify: cost tracking is based on SDK/integration estimates unless actual provider reconciliation has shipped.
- Label: roadmap items as roadmap, not current product.
Launch Pillars
1. SDK Adoption
The SDK launch path is strongest when a developer can install, instrument, and see value in minutes. Required launch state:- clear installation and quickstart docs;
@guarded_actionexamples for sync and async functions;- policy, guardrail, approval, replay, and integration examples;
- local ledger docs that explain exactly what is recorded;
- remote sync docs with API key setup and shutdown/flush behavior.
/quickstart/sdk/python/installation/sdk/python/instrumentation/sdk/python/policies/sdk/python/guardrails/sdk/python/replay/sdk/python/framework-integrations/sdk/python/llm-integrations
2. Platform Reliability
The platform must make SDK telemetry useful and trustworthy. Required launch state:- authenticated ingest with API keys;
- tenant-scoped runs/actions/ledger APIs;
- policy sync;
- approval APIs and decision history;
- intervention and replay APIs;
- WebSocket updates for console workflows;
- deployment and migration notes for the chosen production path.
/platform/overview/platform/ingest/platform/auth/platform/runs/platform/policies/platform/approvals/platform/interventions/platform/replay/platform/websockets/platform/self-hosting
3. Console Operations
The console should support the operator workflows that make runtime control credible. Required launch state:- quickstart/onboarding path;
- runs explorer;
- intervention review;
- approval inbox;
- policy management;
- kill switch;
- replay hub and determinism analysis;
- agent overview;
- ledger export and audit workflows;
- analytics sufficient to show cost and operational signals.
/console/overview/console/quickstart/console/runs/console/interventions/console/approvals/console/kill-switch/console/replay/console/ledger/console/agents/console/analytics/console/settings
Critical Launch Work
Track active delivery here instead of creating new root-level plan docs.| Area | Outcome | Status |
|---|---|---|
| Claims accuracy | Public docs and marketing match shipped behavior | Active maintenance |
| SDK quickstart | Install, instrument, local ledger, remote sync path is clear | Documented |
| Replay | Replay docs and examples explain cost-free debugging and determinism | Documented |
| Approvals | Console approval flow and API behavior are documented | Documented |
| Notifications | Slack/email approval notification path | Validate shipped state before claiming |
| Compliance exports | Ledger/compliance export path is documented and testable | Validate shipped state before claiming certification |
| Framework integrations | LangChain, CrewAI, AutoGen, LangGraph claims map to code and examples | Validate before each release |
| Production deployment | Railway/Supabase/deployment checklist reflects current hosting choice | Needs periodic review |
Launch Readiness Checklist
- SDK quickstart succeeds from a clean virtual environment.
- Remote sync works against staging or production with a new API key.
- Console onboarding produces at least one visible run.
- Policy enforcement blocks a denied action and records an intervention.
- Approval workflow can be requested, decided, and reflected in the ledger.
- Kill switch blocks the intended scope and is auditable.
- Replay example runs locally and posts or displays a replay summary.
- Ledger export works for a bounded date range.
- Documentation has no top-level stale phase reports or duplicate architecture docs.
- Public claims have an owner and evidence link.
Historical Sources
The launch plan condenses these archived inputs:docs/archive/legacy-plans/EXECUTIVE_SUMMARY.mddocs/archive/legacy-plans/CODEBASE_REVIEW_AND_ROADMAP.mddocs/archive/legacy-plans/IMPLEMENTATION_PRIORITY_MATRIX.mddocs/archive/legacy-plans/IMMEDIATE_ACTIONS.mddocs/archive/legacy-plans/HONEST_CLAIMS_AUDIT.mddocs/archive/legacy-plans/PHASE_7_SUMMARY.md