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 agentmods add commands/josephscript/claude-speak/speak-limitgit clone --depth 1 https://github.com/JosephScript/claude-speakWhat 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 | $0.00018 | $0.00268 |
| Opus 5 | $0.00009 | $0.00134 |
| Sonnet 5 | $0.00004 | $0.00054 |
| Haiku 4.5 | $0.00002 | $0.00027 |
Grade B, and why
speak-limit scanned grade B with 1 finding 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 2d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **No argument**: report the current limit. Run `cat ~/.claude/claude-speak/limit 2>/dev/null`; if empty/missing, the limit is the default (2000 bytes). What it actually says
Based on the argument passed to this command, do exactly one of the following via Bash, then report the result in one short sentence — nothing else:
- No argument: report the current limit. Run
cat ~/.claude/claude-speak/limit 2>/dev/null; if empty/missing, the limit is the default (2000 bytes). - A positive integer (e.g.
1200): set that as the new limit:mkdir -p ~/.claude/claude-speak && echo <number> > ~/.claude/claude-speak/limit. reset: clear back to the default:rm -f ~/.claude/claude-speak/limit.- Anything else: don't write anything — tell the user the argument wasn't understood (must be a positive integer or
reset).
This limit applies everywhere the mechanical formatter runs: the automatic /toggle-speak Stop hook and /speak (local mode). It has no effect on /speak llm, whose length is controlled by the model's own summarization instead of this cap.
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.
- 2d ago First seen · 14 lines · 18 tokens per session scan B ab2500e2775c
speak-limit is a command published in the GitHub repository JosephScript/claude-speak (3 stars, last pushed 1mo ago), licensed MIT. It adds 18 tokens to every session and 268 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.