agentleFS
HainrixzHainrixz/jarvis-architect/.claude/skills/jarvis-add/SKILL.md

jarvis-add

Skill18 starsChanged 31 days agoMITView on GitHub

Add something to an existing Jarvis — a channel (Telegram, WhatsApp, Discord, voice), a tool or MCP server (Gmail, Calendar, Notion, HubSpot, Facturapi…), a skill, a scheduled job, the autopilot or an L3 action, a taught skill, or audio/video understanding — with the same gates and security controls as the original build. EN — "add WhatsApp to my Jarvis", "connect my calendar", "turn on the autopilot", "teach it a skill", "add video". ES — "agrégale WhatsApp", "conecta mi calendario", "que también haga…", "que avance solo", "activa el piloto", "agrega una acción L3", "enséñale una skill", "agrega video", "que entienda audios".

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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
  1. Very specific:

    Read the relevant capability file and the runtime file's section for it.

    line 21
  2. Very specific:

    "Agrega una acción L3": only a type from the reversible list in autonomy.md § Levels, never for sensitive packs.

    line 31

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<!-- Copyright (c) 2026 Enrique Rocha (@soyenriquerocha) · tododeia.com · id er-c08d5d7d -->

/jarvis-add

Load the jarvis-start skill's rules. This is a mini DEEP DIVE → ARCHITECTURE → BUILD loop for one addition:

  1. Locate the Jarvis (.jarvis/tasks.json → jarvis_dir or ask) and read its jarvis-spec.md.
  2. Read the relevant capability file and the runtime file's section for it. Check knowledge/compatibility.md (e.g. WhatsApp unofficial bridge → consent gate; local Ollama → never on Railway).
  3. For a tool/MCP: official remote MCP (OAuth) > pinned maintained community MCP (read its source first; reject anything that asks to "download a helper" or pipe a script) > generated wrapper > Zapier MCP. Start read-only. Write scopes only on explicit request, behind the approval gate. For the three built-in capabilities:
    • "Que avance solo" / "activa el piloto" (knowledge/capabilities/autonomy.md): from L0 to L1, add the autopilot job and its files (questions/4-construir.md step 11). Never above L1 by default.
    • "Agrega una acción L3": only a type from the reversible list in autonomy.md § Levels, never for sensitive packs. Show the exact line; the owner writes it into AGENTS.md § 10 in their own terminal (IRR-07: Jarvis and you never edit that list), add the autonomy-l3-<type> consent to spec §7, and give the autopilot job only the narrow tool that type needs.
    • "Enséñale una skill" (knowledge/capabilities/learning.md § 3): write the draft in <jarvis>/proposals/skills/<name>/, then bin/promote-skill.sh promote <name> (lint, probation, commit). A skill that sends anything gets disable-model-invocation: true.
    • "Agrega video" / "que entienda audios" (knowledge/capabilities/multimodal.md): questions/4-construir.md step 13 for the missing part; video from links needs the video-urls-terms consent; cloud STT needs its own consent and is refused with a sensitive pack (CMP-39). Not available on cowork.
  4. Show a ≤10-line card: what changes, cost, new risks, consent if needed. Wait for yes.
  5. Append tasks to tasks.json (new ids after the last), run them via the /jarvis-next loop, update jarvis-spec.md §3–§7, re-run jarvis-validator on the whole Jarvis.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.