Agent Baseline

The Agent Accountability Framework is an open governance framework for making AI agents accountable when they hold authority in production. Published and funded by Gravitee, it defines six principles organized around three stages: Assignment establishes an agent's identity, team, and named human sponsor; Alignment defines its role, mission, measurable boundaries, and least-privilege authority; and Administration provides oversight scorecards, guardian agents, and step-by-step recourse when an agent goes off course. The framework is published free for anyone to use, with a concept paper and an invitation for contributors.

Agent S is an open-source computer-use agent framework from Simular. It accepts natural-language tasks, observes a real desktop through screen and accessibility signals, and operates ordinary applications by clicking, typing, scrolling, and executing optional local code rather than relying on per-application APIs or scripts. The project has progressed through three generations: accessibility-tree grounding with an LLM, a vision-based manager-worker architecture using multiple foundation models, and Agent S3, which scales computer-use attempts through parallel rollouts and selection. Its workflow can combine model-driven exploration with reusable automation code, replay saved skills, and recover when the interface changes. The framework supports macOS, Windows, and Linux, works with models from OpenAI, Anthropic, Gemini, OpenRouter, vLLM, and open-weight providers, and is distributed through the `gui-agents` Python package. An optional local coding environment can execute Python and Bash with the current user's permissions, so the repository warns that this capability should be used only in trusted environments.

Keycard is a runtime authorization platform for AI agents. It gives each agent an identity and evaluates the full request context—including the agent, its runtime, the task, and any user it acts for—against policy before issuing credentials. Approved requests receive short-lived, task-scoped tokens for resources such as APIs, data stores, MCP servers, and other agents; denied requests receive no credentials and are logged. Keycard also provides resource configuration, session inspection, attribution-rich audit logs, SIEM export, and immediate credential revocation.

AgentGuard is a web-based scanner for AI-agent Skills, including SKILL.md instructions. It combines deterministic pattern checks with AI-assisted analysis to identify capabilities and risks such as shell-command execution, file-system access, environment-variable and credential access, network requests, external downloads, suspicious URLs, obfuscated instructions, prompt-injection patterns, and destructive operations. It produces a trust report with security and quality scores, a risk rating, detected capabilities, findings, explanations, and recommendations. The scanner is free to use, requires no account, accepts up to 80,000 characters of Skill text, and is designed for analysis before installation.

FrontierAgent is an open-source agent runtime, terminal product, and evaluation suite for long-horizon research and file-based work. Its native terminal TUI provides a ReAct workflow in which one stateful agent researches, reads files, writes deliverables, runs commands, and iterates in a task-scoped sandbox, and an Agent Team workflow in which a coordinator maintains a task board, delegates bounded independent assignments to parallel sub-agents, collects structured reports, and synthesizes the result. Shell and file tools use a shared sandbox with read-only /inputs, working-state /workspace, and persistent /outputs directories. The TUI supports queued instructions during execution, approval-required diffs for mutating operations, local action traces, checkpointed sessions, resumption, and reversion. The same workflow engine powers a subprocess benchmark runner with deterministic artifact collection, concurrency, progress inspection, and reruns of individual failures. The repository also documents connection to the OpenAI-compatible Apodex-1.1 endpoint through the Apodex API.