Configuration reference¶
Exact configuration keys generated from the schema and bundled defaults.
Key families¶
| Family | Contents | Schema paths |
|---|---|---|
| Core and identity | Repository ownership, identity, and machine-wide configuration switches. | 10 |
| Memory and instructions | Project memory, generated instruction templates, and their compatibility aliases. | 10 |
| TUI | Settings for SASE's terminal interface and its configurable panes. | 345 |
| Scheduler | AXE scheduler routines, jobs, and execution limits. | 629 |
| LLM provider | Provider selection, model defaults, and model execution policy. | 103 |
| Agents and runners | Agent launch, completion, capacity, restart, and process-scope settings. | 55 |
| Repositories and version control | Repository relationships, VCS providers, dispatch, commits, and workspace configuration. | 238 |
| Prompts and tools | Macros, named tools, plugins, hooks, and mentor selection. | 108 |
| Automation and work tracking | Beads, goals, gates, SDD, and related automation policy. | 71 |
| Runtime and stored data | Process history, service hosting, artifacts, disk use, and reading surfaces. | 95 |
| Integrations and telemetry | Chat, mobile access, and local telemetry settings. | 51 |
Top-level keys¶
Each link opens its generated entry; nested fields appear on the same family page.
| Config key | Default | Description | Family |
|---|---|---|---|
| ace | {…} (25 keys; values below) | Configuration for sase's TUI | TUI |
| agent_auto_restart | {…} (6 keys; values below) | Controls recovery of agents interrupted by a live update, including quiescence, retry, and storm limits. | Agents and runners |
| agent_hold_confirm_capture_threshold | 10 |
Pending-capture count above which a launch preview's %hold needs interactive confirmation. | Agents and runners |
| agent_hold_default_ttl | "2h" |
TTL applied to sase agent hold create/run when --ttl is omitted. Bare seconds or a number suffixed with s/m/h. |
Agents and runners |
| agent_hold_max_ttl | "12h" |
Maximum TTL accepted by sase agent hold create/run --ttl. Same duration syntax as agent_hold_default_ttl. |
Agents and runners |
| agent_scope_teardown | {enabled, term_grace_seconds, reaper_min_scope_age_seconds, spare_process_patterns} (values below) |
Sweeps leaked processes from agent-owned scopes at runner exit and reaps abandoned scopes. | Agents and runners |
| amd_agents_minimal_template | null |
Deprecated top-level alias for memory.agents_minimal_template. Optional root-relative path to the Jinja2 template used for create-if-missing minimal AGENTS.md files. New configuration must use memory.agents_minimal_template. | Memory and instructions |
| amd_agents_template | null |
Deprecated top-level alias for memory.agents_template. Optional root-relative path to the Jinja2 template used for AMD-managed AGENTS.md files. New configuration must use memory.agents_template. | Memory and instructions |
| artifact_refs | not set | Artifact-reference policies. | Runtime and stored data |
| artifacts | {capture, retention} (values below) |
Bounds automatic artifact capture and controls the opt-in retention policy for stored artifact data. | Runtime and stored data |
| autonomy | {roles} (values below) |
Autonomy profile each generated worker role launches with. | Agents and runners |
| axe | {…} (10 keys; values below) | Configuration for the axe scheduler daemon | Scheduler |
| bead | {…} (5 keys; values below) | Configuration for the bead/issue-tracking subsystem | Automation and work tracking |
| chat_install | {timeout_seconds, restart_attempts} (values below) |
Configuration for chat-driven update workflows | Integrations and telemetry |
| commit | {message} (values below) |
Configuration for SASE commit enforcement. | Repositories and version control |
| commit_hooks | {before, after} (values below) |
Commands to run before VCS dispatch and after a commit-producing dispatch succeeds. | Repositories and version control |
| default_linked_repos | true |
Inject the default plans and beads sidecar repositories plus the intrinsically hidden agents sidecar repository for SASE-managed projects. | Core and identity |
| disk | {pressure} (values below) |
Disk-pressure thresholds and disk-owner diagnostics. | Runtime and stored data |
| dispatch | {…} (7 keys; values below) | Remote dispatch provider and enrolled machine configuration. | Repositories and version control |
| external_mirror | {issues, pull_requests, exclude_labels, pr_authors} (values below) |
Configuration for the external tracker mirror (issues into beads, pull requests into Patches). | Repositories and version control |
| feature_flags | not set | Code-owned SASE feature flags keyed by flag name. | Integrations and telemetry |
| file_hooks | [] |
Non-gating commands matched to repository file events. | Prompts and tools |
| finalizers | {defaults, required, instances} (values below) |
Host-owned finalizer instances selected for SASE agent completion. | Agents and runners |
| gate | {turn} (values below) |
Configuration for durable command-backed gates | Automation and work tracking |
| github_orgs | not set | GitHub organization/user names the user has push access to. Seeds owner completion and gates PR submission; does not choose SSH vs HTTPS (cloning always tries SSH first, then HTTPS) | Core and identity |
| goals | {} |
Goal ledger resolution and publication policy (epic sase-1bu). | Automation and work tracking |
| id | not set | Explicit owner identity. This object is authoritative only in the machine overlay selected by ~/.sase/machine_name; run sase config init to create or repair it. | Core and identity |
| is_sase_managed | false |
Explicitly authorize SASE to manage repository resources, including project memory, root AGENTS.md, and SDD initialization. Only the target repository's own sase/sase.yml is consulted. | Core and identity |
| linked_repos | [] |
Deprecated alias for 'repos.linked'. Still parsed during the compatibility window; prefer 'repos.linked' in new config. | Repositories and version control |
| llm_provider | {…} (9 keys; values below) | Configuration for the LLM provider abstraction layer | LLM provider |
| machine_name | not set | Deprecated migration input for legacy machine overlays. Run sase config init to move it to id.machine_name; new configuration must use the nested identity object. | Core and identity |
| macro_aliases | {c, p} (values below) |
Aliases for macro references, substituted before any other macro processing (e.g., {"#deploy_notes": "#release-notes"}) | Prompts and tools |
| macros | {…} (12 keys; values below) | Reusable text templates that can be referenced in prompts | Prompts and tools |
| managed_tmp | {…} (5 keys; values below) | Retention horizons and pressure thresholds for the managed SASE temp root reaped by sase disk reap / the hourly managed_tmp_reap job. |
Runtime and stored data |
| markdown | {print_width} (values below) |
Markdown prose formatting policy for SASE-generated Markdown. | Runtime and stored data |
| max_agent_pipe_chain | 8 |
Maximum number of sase pipe hops in one session chain. The original agent is depth 0; each successful pipe increments depth by one and is refused when the next link would exceed this bound. | Agents and runners |
| max_running_agents | 10 |
Host-wide runner-capacity budget. Normal launches claim 1.0 capacity unit; %queue(weight=...) / %q(w=...) may request non-negative finite weights (%q(w=0) adds no load). Serial sessions share one live claim, while independent clan members and parallel session members each hold their own claim. | Agents and runners |
| memory | {…} (5 keys; values below) | Configuration for agent memory and generated agent-instruction files (AGENTS.md and its provider shims). | Memory and instructions |
| memory_readme_template | null |
Deprecated top-level alias for memory.readme_template. Optional root-relative path to the Jinja2 template used for generated sase/memory/README.md files. New configuration must use memory.readme_template. | Memory and instructions |
| memory_sase_template | null |
Deprecated top-level alias for memory.sase_template. Optional root-relative path to the Jinja2 template used for generated sase/memory/sase.md files. New configuration must use memory.sase_template. | Memory and instructions |
| mentor_profiles | not set | Mentor profiles define which mentors to apply to which files | Prompts and tools |
| metahooks | not set | Metahooks intercept failing hooks before the summarize agent runs | Prompts and tools |
| mobile_gateway | {…} (13 keys; values below) | Configuration for the workstation-hosted mobile gateway | Integrations and telemetry |
| monitor | {tool_wrap, evidence_limits} (values below) |
Detached monitor and continuation settings. | Runtime and stored data |
| pager | {syntax} (values below) |
Standalone and embedded SASE pager reading-surface policy. | Runtime and stored data |
| plugins | {required} (values below) |
Project-declared plugin dependencies. A linked or sidecar checkout is not an install; each entry is checked against distributions in the running environment. | Prompts and tools |
| procs | {history_limit, runtime_orphan_horizon_seconds, runtime_orphan_max_removals} (values below) |
Retention limits for finished proc records and their orphaned runtime directories. | Runtime and stored data |
| repos | {} |
Repositories related to the current project. | Repositories and version control |
| runner_slots | {deference_seconds_per_step, deference_max_seconds} (values below) |
Priority-based deference for runner-slot waiters, bounded by a maximum delay. | Agents and runners |
| sdd | {bead_refresh, repo, bead_push_log_retention, push_after_commit} (values below) |
Configuration for Structured Design Documents | Automation and work tracking |
| service | {procs} (values below) |
Machine-level service procs the sase service host starts, restarts, and stops. Project-local sase.yml entries are ignored; the host resolves defaults, plugin, user, and machine-overlay layers only. | Runtime and stored data |
| sibling_repos | [] |
Deprecated alias for 'repos.linked'. Still parsed during the compatibility window; prefer 'repos.linked' in new config. | Repositories and version control |
| tasks | not set | Deprecated alias for 'procs'. Still parsed during the compatibility window; prefer 'procs' in new config. | Runtime and stored data |
| telegram | {commands} (values below) |
Configuration for the SASE Telegram integration. | Integrations and telemetry |
| telemetry | {enabled, flush_interval_seconds, retention, health_thresholds} (values below) |
Local telemetry settings | Integrations and telemetry |
| timezone | "" |
IANA timezone name governing all SASE wall-clock display and timestamp generation (e.g., 'America/New_York', 'US/Pacific'). Defaults to the system timezone when unset. | Core and identity |
| tmux_agent | {…} (6 keys; values below) | Configuration for launching interactive agent CLIs in a new tmux window | Agents and runners |
| tool_runs | {…} (7 keys; values below) | Operational ToolRun ledger retention policy. Uses ordinary config-layer precedence, independent of the project-owned tools catalog. | Runtime and stored data |
| tools | not set | Project-owned named tool catalog. Read only from the project's sase/sase.yml as complete entries; user, machine, plugin, and builtin layers cannot change execution identity. | Prompts and tools |
| update | {dev_root} (values below) |
Configuration for SASE self-update and install-mode switching | Runtime and stored data |
| use_chezmoi | false |
Remap home macro paths to chezmoi-managed equivalents under ~/.local/share/chezmoi/home/ | Core and identity |
| vcs_provider | {provider, use_project_pr_prefix, pr_tags} (values below) |
Configuration for the VCS provider abstraction layer | Repositories and version control |
| vcs_ref_completion | {enabled} (values below) |
Project, Patch, and namespace completion at the root of VCS workflow refs such as #gh:. | Repositories and version control |
| vcs_repo_completion | {enabled, cache_ttl_seconds, max_repos} (values below) |
Repository-name completion inside VCS workflow refs such as #gh:owner/. | Repositories and version control |
| workspace | {…} (5 keys; values below) | Workspace directory layout policy (Phase 1+: sase-3p WorkspaceStore). | Repositories and version control |
Behavior pages¶
These pages describe configuration behavior that schema values alone cannot express. The generated family pages remain the source for key paths, types, defaults, and allowed values.
| Subject | Behavior |
|---|---|
| Files and merging | File locations, owner identity, merge precedence, and directory layout. |
| Core behavior | Identity, generated instructions, and machine-wide settings. |
| ACE behavior | Behavior settings for SASE's terminal interface. |
| ACE keymaps | Configurable keymaps for Admin Center and other TUI surfaces. |
| ACE prompts | Prompt completion, snippets, history, and agent decks. |
| Artifacts | Agent artifacts and artifact references. |
| LLM provider behavior | Provider selection, usage limits, retries, and execution budgets. |
| Finalizers | Commit hooks and post-run finalizers. |
| Monitor behavior | Monitor wrappers and continuation settings. |
| Repositories | Repository lists, dispatch, VCS, and completion settings. |
| Scheduler behavior | AXE scheduler settings and runtime behavior. |
| Prompts and tools | Hooks, plugins, mentors, macros, and named tools. |
| Agents and runners | Agent admission, capacity, and lifecycle behavior. |
| Runtime and storage | Processes, services, local storage, and telemetry. |
| Display and reading | Markdown rendering, paging, and reading surfaces. |
| Integrations | Chat and mobile integration settings. |
| Automation | SDD, beads, and external issue mirroring. |
| Feature flags | Feature flag behavior and persisted state. |
| Workspaces | Workspace directory layout and lifecycle. |
Description coverage¶
670 schema paths need descriptions. Descriptions are maintained
in src/sase/config/sase.schema.json.
Related¶
- Configure the essentials — common setup values.
- Configuration file and merge behavior — files and layer precedence.
_[project]: A sase project is a named unit of work registered with SASE. _[workspace]: A sase workspace is a numbered clone of a project's primary repo, managed by the workspace store and tracked in that project's registry.json. _[agent]: A sase agent is an agent session or a single agent that does not belong to a session. _[prompt]: A prompt is the text SASE hands to a provider to start an agent turn. _[macro]: A macro is a reusable prompt triggered with #name in agent prompts. _[Patch]: A Patch is SASE's local unit of change. _[bead]: A bead is one tracked work item in SASE's git-native issue tracker. _[plan]: A plan is a persisted, approved artifact recording the intent behind agent work. _[TUI]: The TUI is SASE's terminal interface for running agents and reviewing results. _[epic]: An epic is an executable multi-phase plan run with sase bead work. _[gate]: A sase gate is a durable, command-backed request for one user decision, created when work must pause for the user. _[monitor]: A sase monitor is a monitor turn that runs one long command under a detached supervisor, so the command outlives the agent that started it. *[stitch]: A stitch is the lightweight ordered change record inside a Patch's STITCHES: section.