Contexity exposes tools through MCP (Model Context Protocol) so agent hosts can retrieve and write context without direct access to the local store. These tools are called automatically by hosts like Codex and Claude Code — you do not call them manually during normal work.
These tools manage the lifecycle of a single agent run: starting it, recording events during execution, and closing it out with metrics and checkpoint data.
| Tool | Use |
|---|
context_start_run | Start a host-visible task run |
context_get_run_policy | Read current host contract and policy |
context_record_run_event | Append replayable run events |
context_close_run | Close a run and write metrics/checkpoint data |
context_get_run_report | Read local run report and metrics |
Project Context Tools
These tools give the agent access to project context: attaching a project, bootstrapping a fresh session, retrieving a bounded context pack, and recording or invalidating facts.
| Tool | Use |
|---|
context_attach_project | Attach an initialized project |
context_bootstrap_session | Initialize, attach, and fast-analyze frictionlessly |
context_retrieve_pack | Retrieve bounded task-aware context |
context_append_event | Capture project execution context |
context_invalidate_fact | Mark context stale, superseded, or invalidated |
These tools expose the project graph so the agent can reason about structure, search for relevant code surfaces, and map the impact of changes before and after edits.
| Tool | Use |
|---|
context_analyze_project | Refresh project intelligence |
context_get_project_graph_summary | Read graph and project map summary |
context_search_project | Search project intelligence before broad grep |
context_impact | Map impact for known files or routes |
context_detect_changes | Detect changed files and impact after edits |
context_route_map | List route surfaces |
context_list_project_issues | List intelligence and freshness issues |
External Context Tools
These tools let the agent capture and manage external references — Slack threads, issue links, documentation pages, or inspiration sources — and the signals extracted from them.
| Tool | Use |
|---|
context_capture_external_reference | Capture Slack, issue, docs, repo, or inspiration sources plus extracted signals |
context_list_external_context | List captured references and signals |
context_invalidate_external_context | Invalidate an external reference or signal |
These tools manage global preferences that persist across all projects on the machine: reading verified preferences, storing explicit ones, removing outdated ones, and proposing uncertain candidates for later verification.
| Tool | Use |
|---|
context_get_global_profile | Read active verified global preferences |
context_upsert_global_fact | Store explicit user-directed global preferences |
context_delete_global_fact | Forget/remove global facts |
context_propose_global_fact | Propose uncertain candidates |
context_list_global_candidates | Inspect pending global candidates |
These tools let the host read and update the metrics display policy, controlling whether post-run usage summaries are shown to you in the terminal.
| Tool | Use |
|---|
context_get_metrics_display | Read metrics display policy |
context_set_metrics_display | Enable or disable visible metrics |
MCP tools are intended for hosts. During normal work, you should not need to call them manually.