Scheduler

This page covers automated scheduling and how to recover a scheduler skill after a bad edit.

Scheduler skills (isolated namespace)

Scheduler-triggered skills live in a separate tree, not scanned by host.Scanner():

~/.config/agenvoy/skills/scheduler/<short>-<hash8>/SKILL.md

Aspect	Regular skill	Scheduler skill
Path	`~/.config/agenvoy/skills/<name>/SKILL.md`	`~/.config/agenvoy/skills/scheduler/<short>-<hash8>/SKILL.md`
Frontmatter `name`	`<name>`	`<short>-<hash8>` (no `scheduler-` prefix)
`/<name>` autocompletion	yes	no — surfaced as `/sched-<name>` (warn-purple) at the bottom of the picker
Trigger	`MatchSkillCall` then run_skill	(a) cron / one-shot fire from daemon `runtime.SetRunner`, (b) manual `/sched-<name>` from TUI

Creation flow

The scheduler-skill-creator skill is the canonical entry for new schedules. It:

Pre-flight gate (Step 0): if the user message lacks a time token (+5m / HH:MM / etc.) or task token, it must call ask_user first — no defaulting to +10m, no inferring "probably 9am".
Runs python3 scripts/init_scheduler_skill.py <short> to create the skill dir with hash suffix.
Patches the SKILL.md body (description + task + output format).
Calls add_schedule to bind the schedule.

Direct add_schedule calls are allowed only for rebinding existing schedules (changing the time of an already-created scheduler skill).

Manual execution (`/sched-<name>`)

The TUI command picker lists every directory under scheduler/ as /sched-<name>. Selecting one reads the body and dispatches it to the current agent with a preamble that blocks weaker models from misreading the SKILL.md-shaped body as a schedule-creation request and re-running the creator.

The preamble enforces:

Execute the existing scheduler skill immediately and output results
Do not activate scheduler-skill-creator
Do not run init_scheduler_skill.py
Do not call add_schedule

Daemon fire

runtime.SetRunner registers runSkill(ctx, sessionID, skillName). When the scheduler fires (cron tick or one-shot deadline), the runner:

Reads body via filesystem.ScheduleSkillBody(skillName).
Ensures the session directory exists; writes a default bot.md.
Runs the body through exec.ExecWithSubagent — an in-process subagent with always-allow context.

One-shot tasks are removed and the skill dir is trashed after a successful fire.

Recovering from a bad edit

Automatic post-run skill rewriting was removed — a failed run no longer edits the skill behind your back. Skill and tool files are still git-versioned under ~/.config/agenvoy/, and every write_skill / patch_skill / remove_skill auto-commits, so recovery is explicit:

Step	Tool
Inspect the revision history of skill or tool storage	`list_revisions`
Hard-reset that storage to an earlier revision	`restore_revision`

Both are scoped to ~/.config/agenvoy and never touch the user's project repository. A removed skill lands in .Trash/ and stays recoverable the same way.