Macro directives¶
The TUI, macro language server, and launch parser share this Rust-owned contract. Aliases, syntax forms, argument summaries, and descriptions below are generated from that contract. The marker directive uses its current macros_enabled spelling here; %xprompts_enabled remains accepted for old prompts.
| Directive | Alias | Forms | Arguments | Purpose |
|---|---|---|---|---|
%model |
%m |
:value, (...) | :model or (model, alias=model) | Override the LLM model for this prompt |
%effort |
%e |
:value | :level | Set the reasoning-effort level for this prompt |
%final |
— | :value, (...) | :instance|!instance|none or (instance, ...) | Select configured finalizer instances for this launch |
%id |
%i |
:value, (...), bare | :agent-id or :name.{@key}; ([id], bead=, clan=/session=/tribe=) | Assign an agent ID with optional bead, clan, session, or user-managed tribe |
%clan |
%c |
:value, (...) | :name or :name.{@key}, (name, tribe=/summary=/summary_script=), or :name:: summary | Declare a new parallel agent clan |
%wait |
%w |
:value, (...), bare | :agent or (agent, bead=, hood=, time=, for_epic=) | Wait for another agent/workflow and/or a time floor |
%queue |
%q |
:value, (...) | :N or |
Set this launch's capacity budget, |
%hold |
— | :value, (...) | :agent or (agent, pending, future, hood=, tribe=, ttl=, scope=) | Hold selected pre-run agents until this launch settles |
%dispatch |
— | :value, (...) | :machine or (machine) | Send this launch to an enrolled remote machine |
%tab |
— | :value, (...) | :name or (name) | Place this launch's presentation root on a named agent tab |
%if |
— | (...), :: body | %if(should_run=true|false) omits disabled segments; %if:: gates admission with a script guard | Statically omit a prompt segment, or gate launch with a typed predicate |
%proc |
— | (...), :: body | %proc("cmd"), %proc(bash=|python=), or %proc:: plus one fence | Launch a stand-alone proc unit with a Bash or Python body |
%auto |
%a |
:value, bare, + | :mode (plan|tale|epic|manual|off) | Request automatic gate resolution; use %auto, %auto+, or %auto: |
%hide |
%h |
bare, + | flag | Hide the agent from the default Agents tab display |
%repeat |
%r |
:value | :count | Run the prompt multiple times |
%alt |
— | %{…}, :value, (...) | (variants) | Split prompt into variants with different text; shorthand %{A | B} |
%macros_enabled |
%xprompts_enabled (legacy) |
:value | :false|true | Enable or disable macro expansion for a text region |
Argument details¶
%model¶
Override the LLM model for this prompt
%effort¶
Set the reasoning-effort level for this prompt
Suggested values: none, minimal, low, medium, high, xhigh, max.
%final¶
Select configured finalizer instances for this launch
Suggested values: none.
%id¶
Assign an agent ID with optional bead, clan, session, or user-managed tribe
| Keyword | Meaning | Suggested values |
|---|---|---|
bead |
Associate this launch with a bead, including a closed bead; see Bead Claim Lifecycle | — |
clan |
Derive the full ID and join this agent clan | — |
session |
Attach this suffix to an existing agent session | — |
family |
Attach this suffix to an existing agent session | — |
tribe |
Assign this agent to a user-managed tribe | — |
%clan¶
Declare a new parallel agent clan
| Keyword | Meaning | Suggested values |
|---|---|---|
summary |
Attach a Rich-markup summary to this clan | — |
summary_script |
Generate this clan's summary with an executable script | — |
tribe |
Assign this clan to a user-managed tribe | — |
%wait¶
Wait for another agent/workflow and/or a time floor
| Keyword | Meaning | Suggested values |
|---|---|---|
agent |
Wait for an explicit agent identity | — |
bead |
Wait until this bead is closed | — |
for_epic |
Also wait for any epic these agents launch | true, false |
hood |
Wait for current members of this agent hood | — |
proc |
Wait for a proc ID or proc name | — |
time |
Start after a duration or absolute wall-clock time | 5m, 1430 |
unit |
Wait for a logical launch unit ID | — |
%queue¶
Set this launch's capacity budget,
Suggested values: 1, 100, 1.5x.
| Keyword | Meaning | Suggested values |
|---|---|---|
capacity |
This launch's capacity budget, replacing max_running_agents, or |
1, 100, 1.5x |
p |
Alias for priority=; lower values start first | 10, 1 |
priority |
Lower values start first; the default is 10 | 10, 1 |
w |
Alias for weight=; non-negative capacity units claimed by this launch; 0 adds no load | 0, 0.25, 1.0, 2.0 |
weight |
Non-negative capacity units claimed by this launch; 0 adds no load | 0, 0.25, 1.0, 2.0 |
%hold¶
Hold selected pre-run agents until this launch settles
Suggested values: pending, future.
| Keyword | Meaning | Suggested values |
|---|---|---|
hood |
Hold launches in this agent hood | — |
scope |
Choose whether the hold applies to this project or host | project, host |
ttl |
Maximum hold duration | 20m, 1h |
tribe |
Hold launches assigned to this user-managed tribe | — |
%dispatch¶
Send this launch to an enrolled remote machine
%tab¶
Place this launch's presentation root on a named agent tab
%if¶
Statically omit a prompt segment, or gate launch with a typed predicate
| Keyword | Meaning | Suggested values |
|---|---|---|
should_run |
Statically keep or omit this prompt segment | true, false |
%proc¶
Launch a stand-alone proc unit with a Bash or Python body
| Keyword | Meaning | Suggested values |
|---|---|---|
bash |
Bash script body | — |
python |
Python script body | — |
timeout |
Total execution timeout from child start | 20m, 1h |
idle_timeout |
Idle-output timeout from child start | 20m, 1h |
cwd |
Working directory for the proc | — |
workspace |
Whether to acquire an operational workspace lease | true, false |
label |
Descriptive label; never identity | — |
%auto¶
Request automatic gate resolution; use %auto, %auto+, or %auto:
Suggested values: plan, tale, epic, manual, off.
%hide¶
Hide the agent from the default Agents tab display
%repeat¶
Run the prompt multiple times
Suggested values: 2, 3.
%alt¶
Split prompt into variants with different text; shorthand %{A | B}
%macros_enabled¶
Enable or disable macro expansion for a text region
Suggested values: false, true.
_[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.