# Cogensec

> Integrity verification, signed decision provenance and token usage for AI agents.

- [Agentegrity](https://docs.cogensec.com/index.md): Measure whether an AI agent keeps its integrity under adversarial pressure, and record every decision in a verifiable chain.
- [Quickstart](https://docs.cogensec.com/quickstart.md): Instrument an agent, score a profile, or check your install in a few lines.
- [Installation](https://docs.cogensec.com/installation.md): Install the Python library, the TypeScript packages, or the coding-agent plugins.
- [How it works](https://docs.cogensec.com/concepts/how-it-works.md): The three self-securing capabilities, the four layers that check them, how layer results become property scores, and how an adapter turns framework hooks into evaluated events.
- [Scores and actions](https://docs.cogensec.com/concepts/scoring-and-actions.md): How the composite score is weighted, which actions each layer can recommend, what enforcement does with them, how approvals fail closed, and how drift, degradation and recovery move the score.
- [Attestation and decision provenance](https://docs.cogensec.com/concepts/attestation.md): What the attestation chain records, how records are hash-linked and signed, how to verify a chain against a pinned key, and how credential fingerprints ride along.
- [Multi-agent systems](https://docs.cogensec.com/concepts/multi-agent.md): How adapters declare the topology of a team, crew or graph, which events and checks that enables, and how to verify links between agents' chains.
- [Token usage](https://docs.cogensec.com/concepts/token-usage.md): What the token counts mean, where each integration reads them from, how repeats are counted once, and where usage appears in events and the session summary.
- [Integrations](https://docs.cogensec.com/integrations/overview.md): The fourteen framework adapters and two coding-agent hook plugins, what each one can enforce, and the contract they share.
- [Python frameworks](https://docs.cogensec.com/integrations/python.md): Install extras, the init() entry point, and the zero-config module for each of the eight Python framework adapters.
- [TypeScript frameworks](https://docs.cogensec.com/integrations/typescript.md): The seven @agentegrity npm packages: install commands, each package's entry point, runtime requirements and environment variables.
- [Claude Code](https://docs.cogensec.com/integrations/claude-code.md): The Claude Code hook plugin: what it checks, how to install and configure it, how it fails, and how to stream sessions to a console.
- [Codex](https://docs.cogensec.com/integrations/codex.md): The Codex hook plugin: the same runtime as Claude Code on Codex's native hooks, how verdicts differ, and what Codex hooks cannot cover.
- [Exporters](https://docs.cogensec.com/exporters/overview.md): How session data leaves an instrumented agent: the SessionExporter protocol, registration, payloads and the exporters list in the session summary.
- [Stream to Agentegrity Pro](https://docs.cogensec.com/exporters/agentegrity-pro.md): The library side of streaming sessions to Agentegrity Pro: HTTPExporter, the agentegrity pro command, the TypeScript client and what gets sent.
- [OpenTelemetry](https://docs.cogensec.com/exporters/opentelemetry.md): Emit agentegrity sessions as OpenTelemetry traces and metrics with OTelSessionExporter.
- [Webhook and Slack alerts](https://docs.cogensec.com/exporters/alerts.md): Post a short alert to a webhook or Slack channel when an evaluation returns block or escalate.
- [Agentegrity Pro overview](https://docs.cogensec.com/pro/overview.md): What the hosted console does with the data your agents stream, how the console is organized, and which numbers the platform observed versus which the agent reported.
- [Get started with Agentegrity Pro](https://docs.cogensec.com/pro/getting-started.md): Create an account and workspace, sign in, understand the trial, and walk through the Setup & SDK wizard to your first reported run.
- [Ingest tokens](https://docs.cogensec.com/pro/ingest-tokens.md): Create, list and revoke the bearer tokens your agents use to send data to a workspace, and replace a token without losing data.
- [Connect an agent](https://docs.cogensec.com/pro/connect-agents.md): Point Python, TypeScript, Claude Code, Codex and OpenTelemetry agents at a workspace, check the connection, and understand how the console identifies agents and closes sessions.
- [Fleet and agents](https://docs.cogensec.com/pro/fleet-and-agents.md): What the Fleet and Agents screens show, how verdicts, severities, score colors and agent status are computed, and what each number on the Fleet screen counts.
- [Runs and sessions](https://docs.cogensec.com/pro/runs-and-sessions.md): The Runs screen for library sessions and OpenTelemetry traces, cross-run span search, the Sessions screen, the session page and how token usage is shown.
- [Issues and incidents](https://docs.cogensec.com/pro/issues-and-incidents.md): The four failure detectors and what triggers each one, how findings are reviewed, and how incidents open, escalate and close.
- [Threats](https://docs.cogensec.com/pro/threats.md): How the Threats screen groups adversarial detections, how severity is banded, and how each outcome (blocked, held, not enforced, unconfirmed, flagged, logged) is decided.
- [Drift](https://docs.cogensec.com/pro/drift.md): Profile changes recorded at every session start, behavioral baselines and the deviation findings computed against them, default thresholds, and how drift findings are reviewed.
- [Attestations and certificates](https://docs.cogensec.com/pro/attestations.md): When the console issues a signed certificate, what it contains and what the signature does and does not prove, the workspace attestation chain, and how to verify both.
- [Governance and audit log](https://docs.cogensec.com/pro/governance.md): Framework controls judged on your workspace's evidence, the policy decisions your agents made, the signed evidence report, and the hash-chained audit log.
- [Alerts and notifications](https://docs.cogensec.com/pro/alerts.md): Send detector and drift findings to a webhook or an inbox with alert rules, verify signed webhook deliveries, read the delivery log, and choose which notifications reach you.
- [Team, roles and access](https://docs.cogensec.com/pro/team-and-roles.md): Workspaces, the viewer, operator and admin roles and what each can do, inviting and adding members, changing roles, disabling access, and switching workspaces.
- [Plans and limits](https://docs.cogensec.com/pro/plans-and-limits.md): What each plan includes, what happens when a workspace reaches a limit or its trial ends, changing plans, and how retention settings work.
- [Troubleshooting](https://docs.cogensec.com/pro/troubleshooting.md): Why a first run does not appear, what each ingest error code means, and why a certificate, chain status or alert is missing.
- [Configuration](https://docs.cogensec.com/reference/configuration.md): Every environment variable the Python library and TypeScript client read, plus the constructor options people most often set.
- [CLI](https://docs.cogensec.com/reference/cli.md): Every agentegrity command: info, doctor, pro, hook, hook-daemon, verify-decisions and report.
- [Events and wire format](https://docs.cogensec.com/reference/events.md): The three exporter payloads, every event type, the session summary and the HTTP API that carries them.
- [Specification](https://docs.cogensec.com/reference/specification.md): The open Agentegrity specification and the project documents that sit next to it, with a link to each.
- [What leaves your process](https://docs.cogensec.com/security/data-flows.md): Every outbound path in the Agentegrity library, what each one carries and how to turn it on or off.
- [Telemetry](https://docs.cogensec.com/telemetry.md): Every anonymous telemetry event the library sends, and how to turn it off.
- [Threat model](https://docs.cogensec.com/security/threat-model.md): What an attacker can do to Agentegrity itself, which mitigations exist and which risks are left to you.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.