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 instructions/assemblyai/blurt/agents-mdgit clone --depth 1 https://github.com/AssemblyAI/blurtWhat 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 | $0.22629 | $0.22629 |
| Opus 5 | $0.11314 | $0.11314 |
| Sonnet 5 | $0.04526 | $0.04526 |
| Haiku 4.5 | $0.02263 | $0.02263 |
Grade A, and why
blurt AGENTS.md 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 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.
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 — 1,172 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
The canonical agent guide for this repository. CLAUDE.md is a shim that points here; it also maps
the Claude-Code-specific tooling under .claude/ (hooks, skills, subagents).
Start here
Blurt is a macOS dictation app powered by AssemblyAI. Tap or hold a
trigger key, speak, and polished text is pasted into the focused app. Transcription is one remote
AssemblyAI dictation API call: the audio plus a request for the service's server-side LLM cleanup
rewrite (config.llm), so the text that comes back is already polished. The user supplies their own
API key. The request's conversation_context carries two context signals, as ordered turns: the
user's recent dictations, then the text immediately before the cursor. The frontmost app, the
window title, the focused field's label, and the selected text are captured for local work (paste
spacing, the injector's window identity, the developer-mode log) and none of them go on the wire.
There is no config.prompt — the context field replaced it. The user's key terms ride a second
field, word_boost, as a flat list of strings. See
Conversation context and key terms.
Four reflexes before you touch anything:
scripts/check.shis the definition of green — the same script CI runs. Off-Mac, onlyscripts/check.sh --portableis available and it is not green in the CI sense. See Verifying your work.- Read Settled decisions before changing architecture. Every entry was tried the other way and reverted; reintroducing one is the most likely way to make things worse.
- Never hand-edit
App/Blurt/Blurt.xcodeproj/project.pbxproj— it is generated fromApp/Blurt/project.yml. Edit the yml, runxcodegen generateon a Mac, commit the result. - No external SPM dependencies in the engine (rule, guarded by
check.sh); the app carries only the localBlurtEnginepackage.
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 · 1,172 lines · 22,629 tokens per session scan A cf9c5b6e9e95
blurt AGENTS.md is an instructions file published in the GitHub repository AssemblyAI/blurt (5 stars, last pushed 4d ago), licensed MIT. It adds 22,629 tokens to every session, about $0.1131 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 2026-08-31.
Other instructions, from other repositories
Applite CLAUDE.md
Instructions for milanvarady/Applite, covering claude.md, project overview, build & run, architecture and data flow.
os-clovy AGENTS.md
Instructions for open-software-network/os-clovy, covering clovy — agent instructions, project, os platform (shared brain), structure and domain & decisions — read before writing code.
scarf AGENTS.md
Instructions for awizemann/scarf: This repo carries its own agent memory — plain files in git, served by the in-repo memophant-mcp MCP server, managed by the Memophant macOS app. It works the same for every agent (Claude Code, Codex, Cursor, Gemini, Copilot). This block is regenerated between the memophant markers…
scarf copilot-instructions.md
Instructions for awizemann/scarf: File memory notes under one of six folders (architecture/conventions/decisions/operations/project/roadmap), never the root. When a note is grounded in code, pass sourcepaths (the repo files it depends on) so Memory Health can drift-check it — an unanchored code note can't be kept…
SkillDeck CLAUDE.md
Instructions for crossoverJie/SkillDeck, covering claude.md, code style: comments required, build & run commands, architecture and key data patterns.
mectrics AGENTS.md
Instructions for farukkamcici/mectrics, covering agents.md — working agreement for mectrics, 0. golden rule: english-only repository, 1. project shape, 2. internationalization (i18n) and 3. menu bar rendering rules.