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/vibeknow/cli/agents-mdgit clone --depth 1 https://github.com/vibeknow/cliWhat 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 | $0.05873 | $0.05873 |
| Opus 5 | $0.02936 | $0.02936 |
| Sonnet 5 | $0.01175 | $0.01175 |
| Haiku 4.5 | $0.00587 | $0.00587 |
Grade A, and why
cli 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 — 484 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent guidance — vibeknow-cli repo
This repo is a Cobra-based Go CLI with an npm distribution wrapper.
Where to look
cmd/— Cobra command definitions, one file per command.internal/— Reusable framework packages (output, i18n, credential, config).docs/superpowers/specs/— Design documents;docs/superpowers/plans/— Implementation plans.skills/— AI agent Skills (empty in P0).
Conventions
- All user-visible strings go through
internal/i18n. - Tests live alongside source (
*_test.go) plustests/integration/. --outputvalues:text(default in TTY),json,ndjson.- Exit codes: see spec §5.4.
Before making changes
- Run
make testandmake build. - Never commit credentials, cache files (
~/.cache/vibeknow/*), ordist/.
Two-stage video workflow
The video pipeline has two stages — agents should understand both:
- Preview (produced by
vk create): a shareable HTML page. Theshare_urlin the output is the primary artifact; it plays the finished video. - Export (optional, via
vk video export): renders a downloadable MP4. Takes several minutes and extra credits.
Commands:
vk create --from <src>— pipeline → preview →share_url.<src>is a path, a URL, a doc_id, or-to read the source text from stdin;--text <string>passes that text directly, for a passage pasted into a conversation rather than saved to a file.vk video wait [task_id] [--session-id <sess>] --for 90s --output json— watch the run for that long, then report the stage it reached and exit 6 withreason: "wait_budget_expired". For callers whose own timeout is shorter than a render; run it again to keep waiting.vk video status [task_id] [--session-id <sess>] --output json— full snapshot:{preview, export, next_actions}. Agents should follownext_actions[].commandto plan the next step.--preview-diralso writes whatever artifacts exist to disk, which is the only route to them for a caller that did not start the run.vk video export [task_id] [--session-id <sess>] --yes --output json— render MP4.--yesskips the confirmation prompt;VIBEKNOW_ASSUME_YES=1works at the environment level.vk video download [task_id] [--session-id <sess>] --dest out.mp4— pure download (requires export to have completed). The file path is--dest;--outputmeans format here as it does everywhere else.
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 · 484 lines · 5,873 tokens per session scan A 99d990f8eb6f
cli AGENTS.md is an instructions file published in the GitHub repository vibeknow/cli (6 stars, last pushed 5d ago), licensed MIT. It adds 5,873 tokens to every session, about $0.0294 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
FastVideo AGENTS.md
Instructions for hao-ai-lab/FastVideo, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
openstory CLAUDE.md
Instructions for openstory-so/openstory, covering claude.md, commands, dev, quality and tests.
open-video AGENTS.md
Instructions for open-video-ai/open-video, covering agents — openvideo product repo, read first, rules (public-safe) and env (before gpu work).
marketing-studio CLAUDE.md
Instructions for ucsandman/marketing-studio: Remotion (studio/) renders all final video for ALL products/brands; assets are copied out to the product's repo at the end. brands/ .json holds per-product tokens (zod-validated via studio/src/lib/brand.ts; mark components in studio/src/brands/marks.ts); templates resolve…
minimax-h3-opencode-skills AGENTS.md
Instructions for unknowlei/minimax-h3-opencode-skills, covering ai installation contract, install target, required installation and safety.
Omni-Rewriter AGENTS.md
Instructions for WayneJin0918/Omni-Rewriter: Maintainer and coding-agent conventions live in docs/AGENTS.md. Also see .cursor/skills/omni-rewriter-h3-pe/SKILL.md, .cursor/skills/omni-rewriter-model-contribution/SKILL.md, and .cursor/skills/omni-rewriter-promo-pipeline/SKILL.md.