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 semantic-craft/iOS-vibebuddy --skill vibebuddy-handoffgit clone --depth 1 https://github.com/semantic-craft/iOS-vibebuddyWrote 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/semantic-craft/ios-vibebuddy/vibebuddy-handoff)<a href="https://agentmods.dev/skills/semantic-craft/ios-vibebuddy/vibebuddy-handoff"><img src="https://agentmods.dev/badge/skills/semantic-craft/ios-vibebuddy/vibebuddy-handoff/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/semantic-craft/ios-vibebuddy/vibebuddy-handoff"><img src="https://agentmods.dev/badge/skills/semantic-craft/ios-vibebuddy/vibebuddy-handoff.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 | $0.00035 | $0.01055 |
| Opus 5 | $0.00017 | $0.00528 |
| Sonnet 5 | $0.00007 | $0.00211 |
| Haiku 4.5 | $0.00003 | $0.00105 |
Grade A, and why
vibebuddy-handoff 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 yesterday.
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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hand off a VibeBuddy task
Read .agents/skills/handoff/SKILL.md and apply its writing discipline. This wrapper
adds the project's source identity and placement contract; it does not duplicate
the underlying skill or add a dispatch/start mechanism. If that local dependency
is unavailable, report the missing skill rather than pretending to have applied it.
For this repository, the location below overrides the underlying skill's default
temporary-directory destination. Save a new note under
.scratch/<feature>/handoffs/<yyyy-mm-dd>-<from-agent>-<to-agent>.md and link its
absolute path in the owning ticket's Comments. In a worktree, use the ticket's
established absolute .scratch/<feature>/ root. Do not migrate old notes or
silently overwrite an existing same-day note; use a distinguishing agent/session
suffix in <from-agent> when needed.
Before writing, get the recorded facts for your own session (ADR-0023). Determine your native ID as described below, then run:
vibebuddy-mcp facts '<agent>:<your-native-id>' --cwd "$PWD"
(or the MCP tool vibebuddy_handoff_facts). Paste its output at the very top of
the note, unchanged: its first four lines are the header below, and its
## Facts (recorded by VibeBuddy, …) section follows. Fill in Ticket: yourself;
the facts list the .scratch/<effort>/ directories the session touched as
candidates. If your ID is not available, write the header by hand with
Source session: unknown and skip facts; if the tool answers "Not recorded",
keep its header and say so in Verification evidence. Never run facts for
another session's key to stand in for your own.
The first four lines are:
Source session: <key or unknown>
Ticket: <absolute ticket path>
Branch: <actual branch>
Worktree: <absolute checkout path>
Determine your own native ID from your current runtime: Claude Code's known
hook/statusline session ID or this session's transcript path; Codex's thread ID;
Cursor's conversation ID; Grok's session ID. Prefix a known ID with
claude-code:, codex:, cursor: or grok-build:. If your own ID is not available,
write Source session: unknown. Never infer authorship from the newest session in
this project, a nearby timestamp, another agent's ID or a historical instruction.
When two agents share a checkout, the source remains the writer's own ID.
Use these sections: Goal, Settled decisions, Authorization, Current changes,
Verification evidence, Next steps, Suggested skills. Reference existing ticket,
plan, ADR, commit, diff and evidence paths instead of copying their contents.
Verification evidence has two parts: what the facts block shows ran (commands
and exit codes, files edited, git state) and what you conclude from it. A claim
the facts do not support is [unverified], however sure the session felt.
Separate completed checks from proposed checks; mark uncertain claims
[unverified]. Redact secrets and unnecessary personal information. Never include
a daemon token or credentials. State the next concrete unfinished action and
preserve limits such as user-owned merging or configuration changes.
On the Mac, a session row whose key this note names shows Handoff ready, and
Continue with… on that row prefills a New task for another agent with this
note's path as its first prompt (the person reviews and presses Start). A Codex
receiver started that way may write in this note's .scratch/<feature>/
directory even from an agent worktree whose .scratch links elsewhere; other
receivers, or a receiver started by hand, may find that directory outside their
sandbox and should report the text rather than force a write. This skill still
does not start anything itself.
The receiver first reads this note, then follows
docs/agents/skills/vibebuddy-history/SKILL.md: read the source's saved summary,
inspect coverage and stale status, and use show only for unresolved questions.
No saved summary means stale is not applicable. An unknown or ambiguous key is a
lookup limitation, not an absent saved summary. A summary must not override a
newer handoff. Confirm current files before resuming 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.
- yesterday Changed · +28 lines c5506b07cf97
- 2d ago First seen · 59 lines · 35 tokens per session scan A 576a4f1ebae7
vibebuddy-handoff is a skill published in the GitHub repository semantic-craft/iOS-vibebuddy (88 stars, last pushed today), licensed MIT. It adds 35 tokens to every session and 1,055 once invoked, about $0.0002 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 2026-09-14.
Other skills, from other repositories
linear-app-setup
Create and configure a Linear agent app so the assistant can manage issues, projects, and workflows under its own identity.
benchmark
Run performance + quality benchmarks. ASR reports WER, RTF, process memory, and throughput across engines/variants. Arguments include asr, tts, vad, diarize, asr-quick.
review-pr
Review a pull request for conceptual fit, architecture impact, adversarial failure modes, security risk, docs impact, regression risk, test coverage, and merge readiness. Use when asked to review a PR, check whether a PR is safe to merge, decide if more tests are needed, perform adversarial or security review, or…
factory-mcp
Use the Warp Factory MCP to hand work to a software factory and collaborate with it — bundle local work and send it to the cloud, find factory tasks from a Slack thread / Linear ticket / description, and pull a task down to test or iterate locally and hand it back.
todoist-cli
Manage Todoist tasks, projects, labels, filters, sections, comments, reminders, and workspaces via the td CLI. Use when the user wants to view, create, update, complete, or organize Todoist items, or mentions tasks, inbox, today, upcoming, projects, labels, or filters.
autopilot
Carry a well-scoped GitHub issue through the full dev loop autonomously, stopping at a per-run tier boundary (PR-ready, or merge+deploy for small reversible changes).