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 tobihagemann/turbo --skill apply-turbo-handoffsgit clone --depth 1 https://github.com/tobihagemann/turboWrote 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/tobihagemann/turbo/apply-turbo-handoffs)<a href="https://agentmods.dev/skills/tobihagemann/turbo/apply-turbo-handoffs"><img src="https://agentmods.dev/badge/skills/tobihagemann/turbo/apply-turbo-handoffs/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/tobihagemann/turbo/apply-turbo-handoffs"><img src="https://agentmods.dev/badge/skills/tobihagemann/turbo/apply-turbo-handoffs.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 | $0.00115 | $0.00968 |
| Opus 5.5 | $0.00046 | $0.00387 |
| Sonnet 5.5 | $0.00023 | $0.00194 |
| Haiku 4.5 | $0.00012 | $0.00097 |
Grade A, and why
apply-turbo-handoffs 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 14d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Apply Turbo Handoffs
Task Tracking
At the start, use TaskCreate to create a task for each step:
- Audit the handoffs
- Act on the user's reply
- Apply the approved changes
- Close out the handoffs
- Run
/stage-commit-pushskill - Run
/update-turboskill
Step 1: Audit the Handoffs
Read every file in .turbo/handoff/. Audit the handoffs that propose lessons for turbo files. Name any handoff that instead carries a session's in-flight work, and leave it untouched. When no lesson handoff exists, present "No lesson handoffs in .turbo/handoff/." Mark Steps 2 through 6 completed, then use the TaskList tool and proceed to any remaining task.
Take each handoff's evidence as a record of the incident, and audit its remedy against the file it targets before recommending any of it:
- Gap — confirm the gap still exists in the current text, is not already handled elsewhere in the skill, its references, or
claude/SKILL-CONVENTIONS.md, and was not already rejected by a decision recorded in auto memory or the target's git history. - Diagnosis — ask whether the real finding is that the skill should not have run there, or that nothing was broken, rather than that the skill must handle the case.
- Remedy — check the proposed text against the conventions and against what the target skill's other steps depend on. Prefer generalizing an existing rule over adding one more special case to it.
- Genericity — strip anything tied to the originating project so the rule states the shared shape alone.
- Weight — weigh how often the new rule would fire against how often it would help.
Output a verdict per lesson as text: apply as proposed, apply adjusted (with the adjusted wording or placement), or drop, each with the reasoning that decided it. Name any lesson that should stay Claude-only, as an intentional divergence under claude/CLAUDE.md. Be willing to conclude that none of it should ship.
Close with how to reply: approve the verdicts as they stand, describe what to change, or hold a lesson back for a later session. Then end the turn.
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.
- 14d ago Changed 66c814893f7e
- 19d ago First seen · 61 lines · 115 tokens per session scan A e16ee7ba4481
apply-turbo-handoffs is a skill published in the GitHub repository tobihagemann/turbo (414 stars, last pushed today), licensed MIT. It adds 115 tokens to every session and 968 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-09-30.
Other skills, from other repositories
ape-commit
Defines how to write and format git commit messages. Use this skill whenever the user asks you to write, draft, suggest, or format a git commit message, summarize changes into a commit, or help stage and describe code changes. Trigger even for casual requests like "write me a commit for this" or "what should my commit…
changelog
Use when writing a changelog or release notes. Covers what belongs in one, writing for users rather than for git, semantic versioning, and documenting breaking changes so nobody is surprised. Trigger on "write release notes for v2.3", "update the changelog", "is this a major or minor version bump". Do not use for…
git-workflow
Use for branching, committing, history repair, and release hygiene. Covers atomic commits, rebase versus merge, bisect, reflog recovery, and undoing mistakes safely. Trigger on "undo my last commit", "i messed up a rebase", "squash these commits", "find which commit broke this". Do not use for pipeline and deploy…
git-commit-helper
Intelligent commit message generator that enforces Conventional Commits and links changes to Jira issues for traceability.
glab-gpg-key
A GitLab command for managing GPG keys used to sign commits, which helps verify who created them.
agy-delegate
Delegate coding tasks to the Google Antigravity CLI (agy) only when the user explicitly requests it, while the orchestrator retains review and landing responsibility.