A dashboard command for OpenLoomi Loop, the part of OpenLoomi that tracks connectors, scheduled decisions, and recent runs.

Use it to view Loop status or refresh connectors. Related Loop actions can run a cycle, schedule a decision, or register custom types, channels, and rules.

It gives one place to inspect pending decisions and connection health, including a fresh connector check when needed.

loop — Command by melandlabs

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/melandlabs/openloomi/loop

Clone the repo

git clone --depth 1 https://github.com/melandlabs/openloomi

Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,718 The whole file, excluding the scripts and references it only reads on demand.

Security scan B 2 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00034	$0.01718
Opus 5	$0.00017	$0.00859
Sonnet 5	$0.00007	$0.00344
Haiku 4.5	$0.00003	$0.00172

Measured 2d ago against content hash 8ebd41a181c1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

loop scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Sends data to an external URLmediumData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

allowed-tools: Bash(curl *), Bash(jq *), Bash(cat ~/.openloomi/token *), Bash(base64 -d *)

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

allowed-tools: Bash(curl *), Bash(jq *), Bash(cat ~/.openloomi/token *), Bash(base64 -d *)

plugins/claude/commands/loop.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/openloomi:loop

Thin doorway into the openloomi-loop sub-skill. With no arguments it prints the dashboard snapshot from GET /api/loop/state. To force-refresh the connector snapshot (which the dashboard reads but caches), say "loop refresh", "refresh connectors", "force refresh", "check connections", or pass refresh as the slash argument (/openloomi:loop refresh). For everything else (run a tick, schedule a decision, register a custom type / channel / classifier rule), say "loop tick", "loop schedule", "register loop type", "add loop rule" — the sub-skill's frontmatter triggers on those phrases and walks Claude through the full API.

Base URL: http://localhost:3414 (fallback http://localhost:3515). The skill doc spells out which port is which.

Steps

Read the bearer token (base64-encoded JWT stored at ~/.openloomi/token) and decode it:
```
TOKEN=$(cat ~/.openloomi/token | base64 -d)
```
Decide the mode:
- Refresh mode (user said "refresh connectors" / "force refresh" / "check connections" / passed /openloomi:loop refresh) → go to the Refresh connectors section below, not this one.
- Default dashboard mode (no args, or "loop dashboard" / "loop state") → continue.

Fetch the dashboard:

curl -sS "$BASE/api/loop/state" -H "Authorization: Bearer $TOKEN" | jq .

Print the JSON to the user. Highlight:
- pending — number of decisions waiting for Run / Dry / Dismiss
- connectors — integration health (one entry per signal channel; note: this is the cached snapshot — see Refresh below)
- lastTickAt — when the last signal pull ran
- prefs — current interval / brief time / wrap time / timezone
If pending > 0, suggest the natural-language follow-up:

Say "loop inbox" to list them, or "loop tick" to pull new signals and classify them now.
If any connector shows connected: false with a lastError like "no composio surface reachable" and the user actually has those connectors configured through Composio (check with composio connections list), suggest:

Read the full file on GitHub · 139 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 139 lines · 34 tokens per session scan B 8ebd41a181c1

Subscribe to this mod's changes

loop is a command published in the GitHub repository melandlabs/openloomi (1,020 stars, last pushed 8d ago), licensed Apache-2.0. It adds 34 tokens to every session and 1,718 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Analytics and metrics Data and AI LLM Memory and context agent ai claw harness-engineering llm-model

loop

Getting it into your agent

What it costs to keep this loaded

Grade B, and why

Sends data to an external URLmediumData exfiltration

Makes network callslowCapability

How it starts

/openloomi:loop

Steps

What this file has done since we first saw it

Other commands, from other repositories

planwt

statuswt

finishwt

review-and-fix

takeoff

boarding