Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/joncovington/MEICAgentnpx agentmods add commands/joncovington/meicagent/paper-startWrote 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/commands/joncovington/meicagent/paper-start)<a href="https://agentmods.dev/commands/joncovington/meicagent/paper-start"><img src="https://agentmods.dev/badge/commands/joncovington/meicagent/paper-start/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/commands/joncovington/meicagent/paper-start"><img src="https://agentmods.dev/badge/commands/joncovington/meicagent/paper-start.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 | $0.00000 | $0.00680 |
| Opus 5.5 | $0.00000 | $0.00272 |
| Sonnet 5.5 | $0.00000 | $0.00136 |
| Haiku 4.5 | $0.00000 | $0.00068 |
Grade A, and why
paper-start scanned grade A with 0 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 1mo 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Start the full MEICAgent paper-trading session: streamer and paper-trading loop.
This is the paper-trading counterpart to /meic-start — it starts the same shared DXLink streamer (paper trading marks positions from real market quotes, exactly like live) but runs the isolated /paper-loop instead of the live trading loop. It never touches data/meic_trades.db, never submits a live order, and is not gated by enable_live_trading — see docs/paper-trading.md for the full design.
Step 1 — DXLink Streamer
Check if the streamer is running:
python src/streamer.py --status
If running is false: start it as a hidden background process.
Start-Process python -ArgumentList 'src\streamer.py' -WorkingDirectory $PWD -WindowStyle Hidden
Step 2 — Paper-trading dashboard
Invoke the /paper-dashboard skill to launch (or confirm already running) the paper-trading dashboard at http://localhost:5051, separate from the live dashboard's port 5050.
Step 3 — Paper-trading loop (scheduled task)
Register the unattended paper loop as a Windows scheduled task and fire the first run immediately:
python src/paper_loop.py --install-task
This creates the MEICAgent-PaperLoop task, which runs python src/paper_loop.py --once every 2 minutes. Each run is a short-lived process that reliably completes, self-heals if one fails, no-ops outside market hours (it's time-gated), and persists across sessions — the robust way to run it unattended on Windows (a long-running detached daemon proved fragile against stray console events). Each --once runs the parallel-shadow engine across every configured symbol: marking/exiting open ICs (per-side stops, the settlement-aware force-close cascade with physical-settlement early close + friction, and cash-settled left-to-expire settlement — no profit target) and evaluating new entries per profile. All writes go to data/paper_trades.db; the live account and data/meic_trades.db are never touched.
(For a one-off manual iteration outside the task — e.g. a final force-close pass — run python src/paper_loop.py --once. On non-Windows hosts, run python src/paper_loop.py in a terminal or wire a cron job instead.)
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.
- 1mo ago First seen · 37 lines · 0 tokens per session scan A d735c0e83738
paper-start is a command published in the GitHub repository joncovington/MEICAgent (6 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 680 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
speckit.bug.fix
Apply the remediation from a bug assessment and record what was changed.