> For the complete documentation index, see [llms.txt](https://docs.tryprotege.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.tryprotege.com/glossary.md). # Concepts ## Artifact An **artifact** is a note, attachment, or link that provides background for a task, helping Proteges make better decisions. ## Audit **Audit** shows Protege run history and outcomes so you can confirm what happened. Audit includes fleet-wide runs, Protege-specific history, run detail timelines, tool usage, logs, teaching notes, and analytics. ## Auto Top-Up **Auto top-up** automatically purchases a credit pack when your balance falls below a threshold you configure. ## Chat **Chat** is the feature that lets you talk directly to Hookshot or a Protege inside Slack threads and Asana task comments by mentioning `@hookshot`. ## Connected An integration or trigger source is **connected** when Hookshot can use it for the intended workflow. ## Conversation Binding A **conversation binding** maps an external conversation (Slack thread or Asana task) to a chat session in Hookshot, maintaining context across messages. ## Credit A **credit** is the unit of consumption for Protege runs. Each run consumes credits based on the AI model's token usage. ## Credit Pack A **credit pack** is a one-time purchase that adds credits to your workspace balance. ## Decision In the Work Hub, a **decision** is an approval or recommendation attached to a task. Proteges can request decisions when they need human judgment. ## Dispatching In the Event Feed, **dispatching** means Hookshot is still working through the event and the final run state is not ready yet. ## Drift Reconciliation **Drift reconciliation** is a periodic full sync between the Work Hub and an external provider that catches changes that real-time listeners or polling may have missed. ## Environment Link An **environment link** pairs a production workspace with a sandbox workspace. Each workspace can belong to only one environment pair. The sandbox is used for testing Protege configurations before promoting them to production. ## Event Feed The **Event Feed** shows incoming events and how Hookshot responded to them. ## Outcome An **outcome** is the result of a run. It is the answer to "what happened after the Protege ran?" ## Pending Write A **pending write** is a Protege-authored change that is waiting for human approval before it is applied. Pending writes are created when the write policy requires approval. ## Project A **project** is a named set of tasks in the Work Hub, with optional custom fields, completion rules, and provider sync. ## Promotion **Promotion** is the process of deploying a Protege's configuration from a sandbox workspace to its linked production workspace. Promotion copies behavior settings, tool access, and schedules while preserving production credentials and run history. ## Provider Sync **Provider sync** is the process of synchronizing data between the Work Hub and an external provider like Asana or Jira. This includes importing records and, for providers that support it, pushing outbound changes. ## Ready **Ready** means the integration or Protege has the required setup in place for live use. ## Run A **run** is one execution of a Protege. Runs can be **standard** (event or schedule triggered) or **chat** (triggered by a conversation mention). Run statuses include `queued`, `running`, `success`, `failure`, and `dead_letter`. ## Sandbox Workspace A **sandbox workspace** is a dedicated testing workspace linked to a production workspace. Proteges can be configured and tested in the sandbox, then promoted to production. Team and people administration in the sandbox is inherited from the production workspace. ## Schedule A **schedule** is a time-based way to start a Protege instead of waiting for an external event. ## SE Access **SE Access** (Solutions Engineering Access) is a controlled overlay that allows Protege AI staff to view or edit your workspace configuration for support purposes. ## Setup Blocker A **setup blocker** is anything that still prevents a Protege from being fully ready, such as a missing connection, missing trigger source, or missing required configuration. ## Skill A **skill** is a reusable markdown instruction document (SKILL.md) that can be assigned to a Protege to give it specific behavior rules. Skills live in the Skills Hub and are scoped to a team. ## Skills Hub The **Skills Hub** is the team-scoped library where skills are created, managed, and assigned to Proteges. ## Skipped In the Event Feed, **skipped** means Hookshot saw the event but did not run the Protege for that event. ## System Chat **System Chat** is the built-in Hookshot responder that handles direct @mention conversations without needing a user-created Protege. ## Tool Access **Tool Access** controls which actions a Protege can take after it starts. ## Trigger Access **Trigger Access** controls which providers are allowed to start a Protege. ## Trigger Source A **trigger source** is the live source of events that can start a Protege. Some trigger sources are automatic (Hookshot creates and maintains them for you, such as the Fathom transcript trigger). Others are manual (you create them yourself on the Integrations page). Managed trigger sources display a "Managed by Work Hub project" label when a Work Hub project owns them. ## Unevaluated In the Event Feed, **unevaluated** means the event has not yet been fully processed into a final decision you can rely on. ## Write Policy The **write policy** controls whether a Protege's changes are applied immediately (`autonomous`) or require human approval (`approval_required`). {% hint style="info" %} If you see a term in the UI and are not sure what action to take next, look for that term again in [Event Feed](/event-feed.md), [Audit](/event-feed/audit.md), or [Troubleshooting](/event-feed/troubleshooting.md). Those pages turn labels into actions. {% endhint %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.tryprotege.com/glossary.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.