jarvis-next
Skill18 starsChanged 31 days agoMITView on GitHub
Resume building a Jarvis from .jarvis/tasks.json — picks the next unblocked task, runs it, runs its verify commands, and marks it done only when they all exit 0. EN — "continue", "next step", "resume my build". ES — "sigue", "siguiente paso", "continuamos", "dónde me quedé".
In this repository6 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- jarvis-addSkill
- jarvis-doctorSkill
- jarvis-quickSkill
- jarvis-refreshSkill
- jarvis-startSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Read ./.jarvis/state.json and ./.jarvis/tasks.json.
- Very specific: line 28
After the last task: move to VERIFY (smoke test + jarvis-validator) as in questions/4-construir.md.
- Distinctive: line 30
Never re-plan, never invent tasks, never edit a verify command to make it pass.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- Copyright (c) 2026 Enrique Rocha (@soyenriquerocha) · tododeia.com · id er-c08d5d7d -->/jarvis-next — continue the build
Load the jarvis-start skill's rules first (keys, money and safe-install rules apply to every task).
- Read
./.jarvis/state.jsonand./.jarvis/tasks.json. Missing → say there's no build in progress and offer/jarvis-start. - Restate the RUNNING BRIEF from
state.json → briefin ≤5 lines. - Flags:
--list→ printNN · status · titlefor every task (👤 foruser_action) and stop.--task <id>→ work on that task if its dependencies aredone; otherwise say which aren't.--done <id>→ run that task's full verify list; markdoneonly if every command exits 0.
- No flag:
- A task left
in_progress= crash recovery. Re-run its verify list. All pass → markdone(commit + tag). Otherwise continue that task. - Else pick the first
pendingtask (array order) whose dependencies are alldone.
- A task left
- Execute it per
questions/4-construir.mdStep 2. One task per invocation unless the person says "sigue con todo" — then loop, pausing at everyuser_actionand every task with acost. - After the last task: move to VERIFY (smoke test +
jarvis-validator) as inquestions/4-construir.md.
Never re-plan, never invent tasks, never edit a verify command to make it pass. If the plan is wrong, say so and offer to go back to ARCHITECTURE for that slot.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

