Summon

Summon

Overview

Summon is the egregore orchestration loop. It reads the manifest (.egregore/manifest.json), selects the next active work item, maps the current pipeline step to a specialist skill, and invokes that skill. After each step it advances the pipeline, checks context and token budgets, and repeats until all items are completed or the budget is exhausted.

The orchestrator never re-implements phase logic. Each pipeline step delegates to an existing skill via Skill() calls. Summon only manages state transitions, retries, and budget guards.

When To Use

• Processing one or more work items through the full intake-build-quality-ship pipeline.
• Resuming an interrupted egregore session (manifest already exists with active items).
• Running autonomously under a watchdog that relaunches on exit.

When NOT To Use

• Running a single skill in isolation (call the skill directly instead).
• Exploratory work where the pipeline does not apply.
• When human review is needed before every step (use manual skill invocations).

Launching the Orchestrator

Always launch the orchestrator agent in the FOREGROUND. Do not use run_in_background: true. The main session becomes the egregore: it blocks on the orchestrator agent until the egregore finishes or is dismissed.

Agent(
  subagent_type: "egregore:orchestrator",
  prompt: "<context about work items and current state>",
  run_in_background: false   // Required
)

If you launch the orchestrator in the background, the main session will have nothing to do and will stop. This defeats the entire purpose of the egregore. The stop hook cannot prevent this because background agents are detached.

The unattended night run

The orchestrator above is an agent inside a session. The night run is the other entry point: a plain process a watchdog can restart, for hours no session is guaranteed to last. It walks one handed-off item (gate, walk, proof). It exits with the gate's own code (1 to 4) when the item is refused and nothing ran, and with 5 when the walk broke after it started:

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/night_run.py" \
  --item-dir .egregore/handoff/<item> --root .

Nothing inside a session starts it. Run it by hand or from a scheduler you configure. The watchdog from /egregore:install-watchdog does not start it: that watchdog relaunches a claude -p session that invokes this skill. See modules/budget.md for why the cooldown path differs.

Manifest Mode

Before launching the orchestrator, ensure the manifest has the correct run mode:

• Default (no --bounded flag): set "indefinite": true in the manifest. The egregore will scan for new work after completing all items and run until dismissed.
• With --bounded flag: set "mode": "bounded" in the manifest. The egregore stops after all items are completed or failed.

If the manifest already exists and has "mode": "bounded" but the user did NOT pass --bounded, update the manifest to "indefinite": true before launching.

After launching, do NOT produce any summary, status table, or "what's happening" output. The orchestrator IS the session now. Let it run.

Orchestration Loop

Follow these steps exactly. Each iteration processes one pipeline step for one work item.

1. Load state

manifest  = Read(".egregore/manifest.json")
config    = Read(".egregore/config.json")
budget    = Read(".egregore/budget.json")

If manifest.json does not exist, stop with an error: "No manifest found. Run egregore init first."

2. Pick the next work item

item = manifest.next_active_item()

If item is None, all work is done. Save the manifest, report completion, and exit.

3. Map current step to a skill

Look up item.pipeline_stage and item.pipeline_step in the Pipeline-to-Skill Mapping table below. Determine the skill name or action to invoke.

4. Invoke the skill

Call Skill() or execute the mapped action. Pass any required context (branch name, issue ref, etc.) from the work item.

5. Handle the result

On success:

• Call manifest.advance(item.id) to move to the next step.
• Reset item.attempts to 0.
• Save the manifest.

On failure:

• Call manifest.fail_current_step(item.id, reason).
• If item.attempts < item.max_attempts, retry the same step on the next iteration.
• If item.status is now "failed", log the failure and move to the next work item.
• Save the manifest.

6. Check context budget

Estimate context window usage. If usage exceeds 80%:

1. Save the manifest to disk.
2. Write a continuation note to .egregore/continuation.json with the current item ID, stage, and step.
3. Invoke Skill(conserve:clear-context).
4. The watchdog or caller will relaunch a fresh session that resumes from the saved state.

7. Check token budget

If the last skill call returned a rate limit error:

1. Record the rate limit in budget.json via budget.record_rate_limit(cooldown_minutes).
2. Save budget.json.
3. Alert the overseer (see notify.py).
4. Schedule in-session recovery (attended sessions only): use CronCreate for a one-shot resume at the cooldown expiry. Its jobs live only in the running session, so this works only when that session is still alive and idle when the window renews.
5. Otherwise exit gracefully, which is the default for an unattended run. The watchdog reads the recorded cooldown and relaunches after the window renews. Call window.plan_resume() rather than choosing by hand.

8. Repeat

Go back to step 2. Continue until all items are completed, all items are failed, or a budget limit is reached.

Pipeline-to-Skill Mapping

The intake stage steps (parse, validate, prioritize) are handled inline by the orchestrator. See modules/intake.md for details.

Delegation Inside the Loop

The build and quality stages delegate execution by default through Skill(conjure:delegation-core). An unattended loop is where the default earns most: nobody is present to notice that an external CLI was available and unused.

The orchestrator does not decide per task. It invokes the mapped skill, and that skill applies the delegation posture with its own Keep Local clauses.

A providers_exhausted result is not a step failure. The orchestrator must not retry the step or mark the work item failed on it. The skill completes the work locally and the pipeline advances. Treating it as a failure would burn the retry budget on a machine where nothing is broken.

To run the egregore with no external models, export CONJURE_DELEGATION=off in the environment that launches it.