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.
npx skills add Deploy-Forward/convoy --skill convoy-nudgegit clone --depth 1 https://github.com/Deploy-Forward/convoyWrote 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/skills/deploy-forward/convoy/convoy-nudge)<a href="https://agentmods.dev/skills/deploy-forward/convoy/convoy-nudge"><img src="https://agentmods.dev/badge/skills/deploy-forward/convoy/convoy-nudge/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/skills/deploy-forward/convoy/convoy-nudge"><img src="https://agentmods.dev/badge/skills/deploy-forward/convoy/convoy-nudge.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 | $0.00070 | $0.00918 |
| Opus 5 | $0.00035 | $0.00459 |
| Sonnet 5 | $0.00014 | $0.00184 |
| Haiku 4.5 | $0.00007 | $0.00092 |
Grade A, and why
convoy-nudge 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 12d 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nudging a deaf neuron
A neuron receives at tool time (PreToolUse), at turn end (the Stop gate), and
when a background inbox --wait completes. It does NOT receive while it sits
idle at its prompt with no background wait running. That is the only gap left,
and this skill closes it from the lead's side. Everything here was run live on
2026-09-05 (docs/briefs/WIDGET.md, slice 5b evidence) and is replicable.
1. Detect from the tape, never by eye
convoy --root <root> inbox --seat <chair> # n = rows waiting
convoy --root <root> feed --since 30m # last row the chair authored
convoy --root <root> panes # is there a live body for it
Stale = rows waiting AND no row authored or drained for longer than the chair's
usual cadence (grok/codex ack within ~1 min when awake). Say which it is:
idle (body alive, nothing waiting), stale (body alive, rows waiting),
gone (no body). Stamp the diagnosis on the thread.
2. Gone: relaunch, scoped
convoy --root <root> relaunch --thread <key> --seat <chair> [--seat ...] --timeout 300
Only the dead chairs. relaunch re-arms each chair's boot prompt with the
token its join minted, queues a "you left off at " inbox row, and counts
connected only from acks stamped after the relaunch. If Windows Terminal kept
the dead pane open ("process exited... Ctrl+D"), close that window first: find
its HWND by class CASCADIA_HOSTING_WINDOW_CLASS and title, PostMessage WM_CLOSE. Stop only processes you can prove are the chair's (pid by name AND
start time equal to the crew launch), by exact pid, never by name.
3. Stale: nudge the pane, title-verified
powershell -File scripts\wt-nudge.ps1 -List # every WT window: hwnd | title
powershell -File scripts\wt-nudge.ps1 -Root <root> -TitleMatch grok -DryRun
powershell -File scripts\wt-nudge.ps1 -Root <root> -TitleMatch grok
The script refuses unless exactly one window matches; takes the foreground
with an Alt tap + AttachThreadInput (a bare SetForegroundWindow is refused by
Windows); moves pane focus with Alt+Arrow and types ONLY into a pane whose
title is the idle title (grok), never one reading "Waiting for response..."
or "Running: ...". The window title is the focused pane's title: that is the
verification. Alt+Left from the leftmost pane moves nothing, so try the other
direction and re-check rather than assume. The prompt is the same for every
pane and self-identifies through whoami, so order cannot misfire. tmux:
select-pane -t <pane> then send-keys -t <pane> '<prompt>' Enter.
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.
- 12d ago First seen · 72 lines · 70 tokens per session scan A cb6191ea6284
convoy-nudge is a skill published in the GitHub repository Deploy-Forward/convoy (0 stars, last pushed yesterday), licensed MIT. It adds 70 tokens to every session and 918 once invoked, about $0.0003 per session on Opus 5. 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 2026-09-05.
Other skills, from other repositories
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.
story-readiness
Validate that a story file is implementation-ready. Checks for embedded GDD requirements, ADR references, engine notes, clear acceptance criteria, and no open design questions. Produces READY / NEEDS WORK / BLOCKED verdict with specific gaps. Use when user says 'is this story ready', 'can I start on this story', 'is…
projects
List all managed projects with status, branch, open PRs, and open issue counts — portfolio-level view.
ring:mapping-feature-relationships
Mapping how features relate and phasing the work: categorizing PRD features, grouping them into domains, charting cross-feature journeys, dependencies, and integration points, and defining the binding Phases that plan.md mirrors one-to-one at Gate 7. Gate 2 of ring:planning-large-features; runs after…