shellm SKILL.md skillA

shellm from laude-institute/headlong is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add laude-institute/headlong/shellm@git:20260819.4ee873e

Pins git:20260819.4ee873e in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
laude-institute/headlong
Path
skills/shellm/SKILL.md
Versions recorded
5
Latest version
git:20260819.4ee873e, committed Aug 19, 2026
First version recorded
May 25, 2026
Size
7,981 bytes
Repository stars
1,269
License
Apache-2.0

Reference for the shellm system — recursive LLM shell, identity management, memory, skills, trajectory, and all CLI tools. Use when working on shellm itself, debugging agent behavior, or understanding how the pieces fit together.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# shellm

> **This skill may be out of date.** The source of truth is always the code in `bin/`. If you find discrepancies, use the `skill-author` skill to update this file and open a PR.

## Architecture overview

shellm is a set of composable bash scripts that turn an LLM into an autonomous agent living in a shell. The stack, bottom to top:

```
llm              raw LLM calls (Anthropic, OpenAI, Gemini)
shellm           recursive execute-in-shell loop on top of llm
traj / context   step log (DAG) + message assembly for multi-turn
mem / skills     persistent memory + learnable capabilities
identity         isolated agent identities (own mem, skills, traj)
think / chat     autonomous thinking + human conversation
focus            goal tracking
```

An agent activates an identity (`source .identities/<name>/activate`), which sets env vars. All tools read from those env vars — no global config files.

## bin/ reference

### Core engine

| Script | Purpose |
|--------|---------|
| `shellm` | Recursive LLM-in-bash loop. Sends a prompt to the LLM, executes returned bash code blocks, feeds output back, repeats until `FINAL` is set. The heart of the system. |
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add laude-institute/headlong/shellm@sha256:ad6ec193260ebbbd

The line at the top of this page pins the label git:20260819.4ee873e, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_hzschjisi53kihrq.svg)](https://markdownregistry.com/a/art_hzschjisi53kihrq)

4 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 5 versions of this file, between May 25, 2026 and Aug 19, 2026, about one every 22 days.

versioncommittedcommitsizeaudit
git:20260819.4ee873e latest2026-08-19 4ee873e 7,981 BA view · diff
git:20260526.6531f0a2026-05-26 6531f0a 7,996 BA view · diff
git:20260525.0fd06ce2026-05-25 0fd06ce 7,993 BA view · diff
git:20260525.93ad7bb2026-05-25 93ad7bb 7,818 BA view · diff
git:20260525.20b6e6a2026-05-25 20b6e6a 7,806 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (7981 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

laude-institute/headlong · 1,269 stars · license Apache-2.0 · pushed 2026-11-01 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_hzschjisi53kihrq
GET https://markdownregistry.com/api/v1/resolve?ref=laude-institute/headlong/shellm
GET https://markdownregistry.com/api/v1/blob/ad6ec193260ebbbdae20d0eff26a6ee63fcf238d6dbe039b0816fdef0d999f65

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from laude-institute/headlong

AGENTS.md AGENTS.md file
laude-institute/headlong · AGENTS.md
git:20261007.d77aadd · audit B · 1,269 stars
CLAUDE.md CLAUDE.md file
laude-institute/headlong · CLAUDE.md
git:20261007.d77aadd · audit B · 1,269 stars
opencode SKILL.md skill
laude-institute/headlong · contrib/opencode/skill/SKILL.md · Delegate a bounded coding task to OpenCode and review its retained candidate, patch, verification, and trajectory…
git:20260925.602d916 · audit A · 1,269 stars
skill-compiler SKILL.md skill
laude-institute/headlong · contrib/skill-compiler/SKILL.md · Generate and run validation cases for a trusted skill, then inspect a saved report of results and assertion provenance.
git:20261023.3e3dcef · audit A · 1,269 stars
chat SKILL.md skill
laude-institute/headlong · skills/chat/SKILL.md · Reply to humans who are chatting with me
git:20260909.6b6b342 · audit A · 1,269 stars
googleworkspace SKILL.md skill
laude-institute/headlong · skills/googleworkspace/SKILL.md · Manage Google Drive, Gmail, Calendar, Sheets, Docs, and Chat via the gws CLI
git:20260509.7ec6622 · audit A · 1,269 stars
mem SKILL.md skill
laude-institute/headlong · skills/mem/SKILL.md · Life context and memory system
git:20261015.494fda1 · audit A · 1,269 stars
obsidian SKILL.md skill
laude-institute/headlong · skills/obsidian/SKILL.md · Read, write, search, and manage notes in an Obsidian vault
git:20260502.9b702aa · audit A · 1,269 stars
skill-author SKILL.md skill
laude-institute/headlong · skills/skill-author/SKILL.md · Create new ShellLM skills. Use when the user asks to write, author, build, or scaffold a new skill, or when an agent…
git:20260820.6323e14 · audit A · 1,269 stars
slack SKILL.md skill
laude-institute/headlong · skills/slack/SKILL.md · Talk with people on Slack — recognize slack-* senders and reach them via chat reply
git:20260914.c9b83aa · audit A · 1,269 stars
telegram SKILL.md skill
laude-institute/headlong · skills/telegram/SKILL.md · Send and receive messages via the Telegram Bot API using curl
git:20260516.4f7c023 · audit A · 1,269 stars
web-research SKILL.md skill
laude-institute/headlong · skills/web-research/SKILL.md · Research topics using the web via curl. Use when the user asks to look something up, find current information, or…
git:20260502.9b702aa · audit A · 1,269 stars

Every file in laude-institute/headlong

Browse by kind, by grade A, or by owner.