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/coven/agents-mdgit clone --depth 1 https://github.com/OpenCoven/covenWhat 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 | $0.01698 | $0.01698 |
| Opus 5 | $0.00849 | $0.00849 |
| Sonnet 5 | $0.00340 | $0.00340 |
| Haiku 4.5 | $0.00170 | $0.00170 |
Grade A, and why
coven 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — coven
Guidance for AI agents (Codex, Claude Code, Hermes, and any Coven familiar)
opening pull requests against this repo. Humans: your canonical guide is
CONTRIBUTING.md — this is the agent-specific layer on top.
Read first:
README.mdfor what this repo is, andCONTRIBUTING.mdfor the full contribution bar (including the "Contributor First 10 Minutes" checkout path).
What this repo is (one line)
Coven is a small, boring Rust authority layer with TypeScript integration packages around it. The development loop must keep that boundary clear: core logic stays in Rust; the npm packages are thin integration surface.
Check, enter a worktree, then claim it
Multiple agent sessions (Codex, Claude Code, familiars) frequently run against the same checkout at once, each in its own worktree. Worktrees keep git operations from racing, but they do not stop two sessions from independently building the same issue — which has happened repeatedly, producing duplicate PRs that a session then has to close. Before you touch code:
-
Check what's already taken. Duplication hides behind divergent branch names — one issue once spawned
fix/output-polish,fix/311-output-polish, andfix/output-polish-311— so branch names alone won't tell you. Check both the shared claim registry and open PRs:coven claim status # active claims, shared across every worktree of this repo gh pr list --state open # is there already a PR for this issue?If the issue is claimed or already has a PR, pick different work or coordinate.
-
Create or enter the task worktree. The automatic fallback identity is worktree-scoped, so enter the worktree before acquiring the claim:
git fetch origin main git worktree add -b <branch> /tmp/coven-<branch-slug> origin/main cd /tmp/coven-<branch-slug> -
Claim it with a shared, issue-keyed token — not your working branch name, which no other session can predict:
coven claim acquire issue-<N> # e.g. issue-311; a TTL-bounded lockClaims live in the repo's shared
--git-common-dir/agent-claims/, so every worktree and session sees them. For long tasks, extend the TTL withcoven claim heartbeat issue-<N>.
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 · 153 lines · 1,698 tokens per session scan A 51468c69173c
coven AGENTS.md is an instructions file published in the GitHub repository OpenCoven/coven (41 stars, last pushed 2d ago), licensed MIT. It adds 1,698 tokens to every session, about $0.0085 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 instructions, from other repositories
helmor AGENTS.md
Instructions for dohooo/helmor, covering agents.md, what is helmor, commands, architecture and three-process model.
claudectl CLAUDE.md
Instructions for mercurialsolo/claudectl, covering claudectl, build & test, architecture, workspace layout and core modules — claudectl-core (crates/claudectl-core/src/).
agents CLAUDE.md
Instructions for eloylp/agents, covering claude.md, project overview, directory structure, config model and build & run.
agiwo AGENTS.md
Instructions for xhwSkhizein/agiwo, covering agents.md, how to use this file, truth source order (adr 0048 / 0049), logging and config hot reload.
Agent-Harness-Kit AGENTS.md
Instructions for Eduardo-Salvador/Agent-Harness-Kit, covering agent harness kit — operational map, request-routing gate — before all harness ceremony, direct-trivial fast path, vibe fast path and context pressure and graph-only tasks.
awesome-agent-infrastructure CLAUDE.md
Instructions for backblaze-labs/awesome-agent-infrastructure, covering claude.md — awesome-agent-infrastructure, what this list is, scope and files, schema, workflow, rules.