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.
git clone --depth 1 https://github.com/oborchers/fractional-ctoWrote 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/oborchers/fractional-cto/baton-pass)<a href="https://agentmods.dev/commands/oborchers/fractional-cto/baton-pass"><img src="https://agentmods.dev/badge/commands/oborchers/fractional-cto/baton-pass/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/oborchers/fractional-cto/baton-pass"><img src="https://agentmods.dev/badge/commands/oborchers/fractional-cto/baton-pass.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 | $0.00026 | $0.01134 |
| Opus 5 | $0.00013 | $0.00567 |
| Sonnet 5 | $0.00005 | $0.00227 |
| Haiku 4.5 | $0.00003 | $0.00113 |
Grade A, and why
baton-pass 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 10d 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Publish a baton signalling that your work is complete, so an agent waiting in a separate process can start its dependent task.
Invoke the agent-baton skill and follow its protocol. This command is a thin wrapper — the skill is the specification.
Step 1: Resolve the baton id
The id comes from $1, or from the instruction you were given at setup ("pass the baton X when done").
Never invent an id. The waiting agent is already listening for a specific string; an id you made up is one nobody is waiting for.
Then, depending on how you got here:
-
The user just typed this command — they are present. If
$1is missing, useAskUserQuestionto ask for the id. If they offer a reused id, remind them of the skill's rule: the id must be unique per chain run, not per task, or it may collide with a leftover baton from an abandoned run. -
You are publishing at the end of an instructed run — the user is almost certainly gone. The id must already be known from your setup instruction. If it is not, do not ask, and do not wait for an answer. A question here blocks forever on someone who left hours ago and strands the downstream agent until its deadline expires. Instead: report loudly that the chain is broken, state which run finished and that no baton could be published, and stop. A visible failure the user finds on their return beats a silent hang.
This is why the skill says to lock the id in the moment you are told, and to echo it back while the user is still watching.
Step 2: Resolve the status
The status comes from $2, and is done or failed.
If not given, determine it from the work you just completed:
- The work succeeded →
done. - The work failed, was abandoned, or you are unsure whether it succeeded →
failed.
Publish a failed baton rather than publishing nothing. An agent that dies silently is indistinguishable from one still working; the waiter learns nothing until its deadline expires. A failed baton tells it immediately.
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.
- 10d ago First seen · 68 lines · 26 tokens per session scan A 55b85b719b58
baton-pass is a command published in the GitHub repository oborchers/fractional-cto (29 stars, last pushed 1mo ago), licensed MIT. It adds 26 tokens to every session and 1,134 once invoked, about $0.0001 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 2026-08-30.
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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
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.