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/punt-labs/biffWrote 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/punt-labs/biff/read-dev)<a href="https://agentmods.dev/commands/punt-labs/biff/read-dev"><img src="https://agentmods.dev/badge/commands/punt-labs/biff/read-dev/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/punt-labs/biff/read-dev"><img src="https://agentmods.dev/badge/commands/punt-labs/biff/read-dev.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 | $0.00006 | $0.03261 |
| Opus 5.5 | $0.00002 | $0.01304 |
| Sonnet 5.5 | $0.00001 | $0.00652 |
| Haiku 4.5 | $0.00001 | $0.00326 |
Grade A, and why
read-dev 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- read — 89% identical, 78 lines differ
How it starts
The opening of the file, as written. The whole thing — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Input
Arguments: $ARGUMENTS
Task
biff is passive/pull. Incoming talk (invites + real-time messages) and /write
mail are held on the biff server and surface by MUTATING the descriptions of the
talk and read_messages tools. This command manages polling and checks for
new activity.
Two complementary delivery paths. plugin/hooks/unread-nudge.sh covers
active sessions: it fires on every UserPromptSubmit and PostToolUse,
reads the same per-session unread file section C reads below, and injects an
additionalContext nudge the moment the count rises — the model sees it on
its very next turn, with no /biff-dev:read-dev invocation required. This
command (the cron loop in section A, and the manual/forced check in section
C) covers the idle gap the nudge hook cannot: a session sitting with no
prompt submitted and no tool called has nothing to fire the hook on, so the
cron's periodic /biff-dev:read-dev invocation is still what surfaces mail
during an idle stretch. Running section C by hand also remains the reliable
way to force-check right now, independent of whether the nudge already
fired.
Argument routing
First, check if $ARGUMENTS matches a polling config command:
- A duration like
1m,2m,5m,10m,15m,30m,1h,2h→ start polling (section A) - Exactly
n,off, orstop→ disable polling (section B) - Exactly
status→ show polling config (section D) - Exactly
force→ force-check now (section C, unconditional)
If none of the above match (empty or unrecognized), fall through to check now (section C).
A. /biff-dev:read-dev <duration> — start polling every <duration>
- Call
mcp__plugin_biff-dev_tty__set_poll_intervalwithintervalset to<duration>— this sets the server-side poll cadence. - Delete every existing
/biff-dev:read-devauto-poll cron job (see "Managing the cron job" below). This must happen before creating the new job, so two back-to-back/biff-dev:read-dev 10mcalls leave exactly one cron job, not two. - Create a new cron job:
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 Changed · +78 lines 16f3a90b3f28
- 1mo ago Changed · +6 lines 5e0459ef9393
- 1mo ago Changed · +122 lines 4eaea17d599f
- 1mo ago First seen · 14 lines · 6 tokens per session scan A 7d285e215a7b
read-dev is a command published in the GitHub repository punt-labs/biff (2 stars, last pushed yesterday), licensed MIT. It adds 6 tokens to every session and 3,261 once invoked, about $0.0000 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 2026-08-31.
Other commands, from other repositories
feedback-to-us
Tell us how the prfaq plugin is working for you (anonymous 1-5 feedback).
setup
Install and configure fuzz, probcli, and lean dependencies.
badge
Generate a stage-colored badge and embed it in your README.
contracts-dev
Generate runtime contracts (preconditions, postconditions, invariants) from a Z specification.
code2model-dev
Create a Z specification for stateful entities in a system.
partition-dev
Derive test cases from Z specification using TTF testing tactics.