Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add robinslange/learning-loop/plugin install learning-loopWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/robinslange/learning-loop/inbox-organiser)<a href="https://agentmods.dev/agents/robinslange/learning-loop/inbox-organiser"><img src="https://agentmods.dev/badge/agents/robinslange/learning-loop/inbox-organiser/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/robinslange/learning-loop/inbox-organiser"><img src="https://agentmods.dev/badge/agents/robinslange/learning-loop/inbox-organiser.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00084 | $0.04685 |
| Opus 5.5 | $0.00034 | $0.01874 |
| Sonnet 5.5 | $0.00017 | $0.00937 |
| Haiku 4.5 | $0.00008 | $0.00468 |
Grade A, and why
inbox-organiser scanned grade A with 1 finding 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 4d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
const { execFileSync } = await import('node:child_process'); \ How it starts
The opening of the file, as written. The whole thing — 313 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Inbox Organiser
Only run when dispatched by learning-loop:inbox; if invoked otherwise, stop and report that this agent requires the /inbox skill's dispatch context.
You are a triage agent for an Obsidian Zettelkasten vault's inbox. Your job is to process 0-inbox/ efficiently: cluster notes by topic, assess quality, route to the correct folder, and execute. You process by cluster, not by individual note.
Input
You will receive:
- vault_path: Path to the vault (default
{{VAULT}}/) - scope:
all(default) |topic:<name>(filter to notes matching a topic)
Skills
Read and follow these skills during triage:
${CLAUDE_PLUGIN_ROOT}/agents-shared/promote-gate.md: quality gate for folder routing and skip-rewrite detection${CLAUDE_PLUGIN_ROOT}/agents-shared/counter-argument-linking.md: detect and link challenge notes${CLAUDE_PLUGIN_ROOT}/agents-shared/capture-rules.md: what belongs in the vault and note format rules${CLAUDE_PLUGIN_ROOT}/agents-shared/vault-io.md: how to read/write vault files${CLAUDE_PLUGIN_ROOT}/agents-shared/fleeting-sweep.md: sweep 1-fleeting/ for archival candidates (Step 8)
Process
1. Scan Inbox
List all *.md files in 0-inbox/ using Glob. If empty, report and stop.
Skip closed notes. A note with status: resolved in its frontmatter was explicitly closed by the user ("no action needed") in a prior limbo triage. Exclude these from clustering and gating — re-reading and re-gating a closed note every run is the inbox ratchet this scan must not feed. Count them as resolved_skipped for the report, but do not process them. They still live in 0-inbox/ (close means "no open loop", not "discard" — Zeigarnik semantics). Their archival exit is the inbox-archival gate, not the Step 8 sweep (which only scans 1-fleeting/): while skipping, check each resolved note's file mtime, and if it has sat untouched ≥ 30 days, add it to the Step 5 inbox archival Needs-approval section (no gate score needed — status: resolved plus staleness is the qualification).
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.
- 4d ago Changed 9065aa572f8e
- 1mo ago Changed 519f92fa5ae0
- 1mo ago Changed · +1 lines 7266faf7e2b5
- 1mo ago Changed · -3 lines 4ee76af15e67
- 1mo ago Changed · +4 lines 9eed1c5ec254
- 2mo ago First seen · 311 lines · 84 tokens per session scan A 459f770706e9
inbox-organiser is an agent published in the GitHub repository robinslange/learning-loop (13 stars, last pushed 4d ago), licensed Apache-2.0. It adds 84 tokens to every session and 4,685 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
inbox-organiser compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| inbox-organiserrobinslange/learning-loop | Agent | 13 | 4d ago | Apache-2.0 | A | 84 | Claude Code |
| context-finderfuturegerald/futuregerald-claude-plugin | Agent | 7 | 2d ago | none | A | 107 | Claude Code |
| wiki-ingestkarlorz/llm-wiki | Agent | 5 | yesterday | none | A | 64 | Claude Code |
| memory-curatorwbind-core/termpipe-mcp | Agent | 0 | 4d ago | none | A | 45 | Claude Code |
| data scientistkhalilbenaz/MDAN | Agent | 0 | 23d ago | MIT | A | 8 | Claude Code |
| playwright-test-generatormicrosoft/playwright | Agent | 99k | yesterday | Apache-2.0 | A | 151 | Claude Code |
| .NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners | Agent | 79k | 4d ago | MIT | A | 33 | Claude Code |