How do I install the Subagent skill?

Run npx skillmds@latest add jasonkneen/subagent in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Subagent safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Subagent?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Subagent free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Subagent?

jasonkneen (@jasonkneen) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Subagent by jasonkneen — Agent Skill

Name: Subagent
Author: jasonkneen

Subagent

The ONLY sanctioned way to delegate work to a lazar sub-agent. Uses locked bin/lazar-spawn — lints the prompt, bounds retries, wall-clock-kills wedged children, and gates success by re-checking declared claims against the filesystem. Prefer pack-and-spawn.sh to write a BRIEF context pack then dispatch. Raw `lazar -p` is vetoed by a pre-tool hook. Use for isolated file-verifiable tasks and parallel fan-out — not for routine bash.

jasonkneen 4887cee 4 files · 14.4 KB Updated Sep 17, 2026

File contents

subagent — how I extend myself

Delegation is a core skill. I never call lazar -p directly (a pre-tool hook vetoes it). I go through the locked spawn surface, preferably with a context pack:

# Recommended: BRIEF.md + thin prompt + lazar-spawn
bash "$LAZAR_HOME/skills/_meta/subagent/pack-and-spawn.sh" <task-id> \
  --objective "..." \
  --out /absolute/deliverable \
  --path /absolute/input \
  --skill create-skill \
  --note "user wants terse output"

# Lower-level (already-written prompt file):
bash "$LAZAR_HOME/bin/lazar-spawn" <task-id> <prompt-file>

bin/lazar-spawn is installed by setup.sh with the same immutability story as bin/lazar. Budget state lives in logs/subagent/. The copy under skills/_meta/subagent/spawn.sh is only a re-exec stub.

When to spawn vs do it yourself

Do it yourself (bash in-process)	Spawn a child
< ~3 tool steps	Isolated unit with a file deliverable
Needs live conversational deixis	Context isolation would help the parent
Exploratory one-off	2+ independent units (parallel via agent-dispatch)
Result is just stdout to reason on	SUCCESS can be checked on disk

Never wrap every bash call as a subagent — latency and token cost explode.

Context pack (enough context without cloning chat)

Children are strangers with a filesystem. Do not dump parent history. Write workspace/tasks/<id>/BRIEF.md (pack-and-spawn does this) containing:

Imperative objective
Absolute input/output paths
SUCCESS: + EXPECT-VERIFY: claims you author
Relevant skill names (not full INDEX)
≤ ~20 lines of prior decisions / constraints
Out-of-scope list
RESULT.md protocol (DONE / BLOCKED)

The spawn -p string stays thin and points at the BRIEF.

Gate on the envelope

lazar-spawn re-checks EXPECT-VERIFY and returns {"status":"ok","matched":true} only if claims hold. Trust matched, not child prose. On matched:false, fix the BRIEF and retry within ceilings.

Bounds (all models)

Kernel stall-nudges (in-child recovery, bounded)
Per-task ceiling LAZAR_SPAWN_CEILING (default 2)
Session budget LAZAR_SPAWN_BUDGET (default 10)
Wall-clock LAZAR_SPAWN_TIMEOUT (default 300s)
Recursion depth cap (kernel MAX_DEPTH)

Self-test

bash skills/_meta/subagent/selftest.sh

Async (optional)

bash "$LAZAR_HOME/bin/lazar-spawn" --detach <task-id> <prompt-file>
# → {"status":"running","job_id":"..."}

bash "$LAZAR_HOME/bin/lazar-spawn" --status <job-id>
bash "$LAZAR_HOME/bin/lazar-spawn" --await <job-id>

Prefer sync spawn for single tasks. Use --detach only for parallel fan-out; still gate on matched via --await/--status.

---name: subagentdescription: The ONLY sanctioned way to delegate work to a lazar sub-agent. Uses locked bin/lazar-spawn — lints the prompt, bounds retries, wall-clock-kills wedged children, and gates success by re-checking declared claims against the filesystem. Prefer pack-and-spawn.sh to write a BRIEF context pack then dispatch. Raw `lazar -p` is vetoed by a pre-tool hook. Use for isolated file-verifiable tasks and parallel fan-out — not for routine bash.---# subagent — how I extend myselfDelegation is a core skill. I never call `lazar -p` directly (a pre-tool hookvetoes it). I go through the **locked** spawn surface, preferably with a**context pack**:```bash# Recommended: BRIEF.md + thin prompt + lazar-spawnbash "$LAZAR_HOME/skills/_meta/subagent/pack-and-spawn.sh" <task-id> \ --objective "..." \ --out /absolute/deliverable \ --path /absolute/input \ --skill create-skill \ --note "user wants terse output"# Lower-level (already-written prompt file):bash "$LAZAR_HOME/bin/lazar-spawn" <task-id> <prompt-file>````bin/lazar-spawn` is installed by `setup.sh` with the same immutability storyas `bin/lazar`. Budget state lives in `logs/subagent/`. The copy under`skills/_meta/subagent/spawn.sh` is **only** a re-exec stub.## When to spawn vs do it yourself| Do it yourself (bash in-process) | Spawn a child ||----------------------------------|---------------|| < ~3 tool steps | Isolated unit with a file deliverable || Needs live conversational deixis | Context isolation would help the parent || Exploratory one-off | 2+ independent units (parallel via agent-dispatch) || Result is just stdout to reason on | SUCCESS can be checked on disk |**Never** wrap every bash call as a subagent — latency and token cost explode.## Context pack (enough context without cloning chat)Children are **strangers with a filesystem**. Do not dump parent history.Write `workspace/tasks/<id>/BRIEF.md` (pack-and-spawn does this) containing:1. Imperative objective2. Absolute input/output paths3. `SUCCESS:` + `EXPECT-VERIFY:` claims **you** author4. Relevant skill names (not full INDEX)5. ≤ ~20 lines of prior decisions / constraints6. Out-of-scope list7. RESULT.md protocol (`DONE` / `BLOCKED`)The spawn `-p` string stays thin and points at the BRIEF.## Gate on the envelope`lazar-spawn` re-checks EXPECT-VERIFY and returns`{"status":"ok","matched":true}` only if claims hold. Trust `matched`, notchild prose. On `matched:false`, fix the BRIEF and retry within ceilings.## Bounds (all models)1. Kernel stall-nudges (in-child recovery, bounded)2. Per-task ceiling `LAZAR_SPAWN_CEILING` (default 2)3. Session budget `LAZAR_SPAWN_BUDGET` (default 10)4. Wall-clock `LAZAR_SPAWN_TIMEOUT` (default 300s)5. Recursion depth cap (kernel `MAX_DEPTH`)## Self-test`bash skills/_meta/subagent/selftest.sh`## Async (optional)```bashbash "$LAZAR_HOME/bin/lazar-spawn" --detach <task-id> <prompt-file># → {"status":"running","job_id":"..."}bash "$LAZAR_HOME/bin/lazar-spawn" --status <job-id>bash "$LAZAR_HOME/bin/lazar-spawn" --await <job-id>```Prefer sync spawn for single tasks. Use `--detach` only for parallel fan-out; still gate on `matched` via --await/--status.

Frequently asked questions