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/opencoven/claude-code-cast/agents-mdgit clone --depth 1 https://github.com/OpenCoven/claude-code-castWhat 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 | $0.00735 | $0.00735 |
| Opus 5 | $0.00367 | $0.00367 |
| Sonnet 5 | $0.00147 | $0.00147 |
| Haiku 4.5 | $0.00073 | $0.00073 |
Grade A, and why
claude-code-cast 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — claude-code-cast
Guidance for AI agents (Codex, Claude Code, and any Coven familiar) opening
pull requests against this repo. This is the agent-specific layer; read
README.md for the full picture.
What this repo is: the
castClaude Code plugin. It emits Coven Agent Event Protocol v0 events from Claude Code hooks so local-first Cast, the Coven daemon, TUI, comux, and future UI surfaces can consume the same agent status stream.
Branch & PR workflow
- Never push to
main. Every change lands via a PR. Branch from currentorigin/main. - Fresh branch per task; use a worktree if multiple sessions may touch this
repo:
git fetch origin main git worktree add -b <branch> /tmp/cccast-<branch> origin/main - Keep the diff scoped to one concern; conventional-commit subjects (
feat:,fix:,docs:,chore:,refactor:). - After merge: delete the remote branch, remove your local worktree/branch.
Before opening the PR
- The hooks are shell scripts — keep them POSIX-portable, quote variables, and fail safe. A hook that errors must not break the host Claude Code session.
- Test hooks end-to-end where possible: trigger the Claude Code hook and confirm a well-formed Agent Event Protocol v0 event is emitted.
shellcheckyour scripts before submitting.
Repo-specific invariants (don't break these)
- Conform to Coven Agent Event Protocol v0. Event shape/field names are a contract consumed by Cast, the daemon, TUI, and comux. Don't rename or drop fields without coordinating the protocol change across consumers.
- Emit, don't block. The plugin observes and reports agent status; it must never gate or slow the host session waiting on a consumer.
- No secrets in emitted events or committed files. Event payloads may travel to multiple local surfaces — keep tokens and private data out.
Attribution — credit contributors correctly
When you re-land or build on someone else's work (a fork PR, an issue author's proposal, a co-author), credit the human contributor with a working GitHub-linked trailer so they appear in the contributors graph and on their profile:
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 · 71 lines · 735 tokens per session scan A 1361fbef442c
claude-code-cast AGENTS.md is an instructions file published in the GitHub repository OpenCoven/claude-code-cast (5 stars, last pushed 1mo ago), licensed MIT. It adds 735 tokens to every session, about $0.0037 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
Lody AGENTS.md
Instructions for LodyAI/Lody, covering repository guidelines, context maintenance, repository boundary, project map and checks and commits.
cicero AGENTS.md
Instructions for 5uck1ess/cicero, covering agent guidance, product boundary, toolchain, verification and reliability and security invariants.
cicero CLAUDE.md
Instructions for 5uck1ess/cicero, a project described as: Self-hosted voice for coding agents. Talk from any browser or a Telegram call, interrupt mid-sentence, clone any voice, and hand real work to agents that ship PRs. Local STT/TTS; real-time voice client for ACP.
cli AGENTS.md
Instructions for konteks/cli, covering agent instructions, project overview, tech stack, repository layout and common commands.
pi-arcweld AGENTS.md
Instructions for appautomaton/pi-arcweld, covering local pi workspace rules, workspace layout, build rules, user pi command and git hygiene for pi-mono.
codex-tg AGENTS.md
Instructions for mideco-tech/codex-tg, covering agents, repository purpose, working mode, agent workflow kernel and context pull map.