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/okikesolutions/guerillaglass/agents-mdgit clone --depth 1 https://github.com/okikeSolutions/guerillaglassWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/okikesolutions/guerillaglass/agents-md)<a href="https://agentmods.dev/instructions/okikesolutions/guerillaglass/agents-md"><img src="https://agentmods.dev/badge/instructions/okikesolutions/guerillaglass/agents-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.01849 | $0.01849 |
| Opus 5 | $0.00924 | $0.00924 |
| Sonnet 5 | $0.00370 | $0.00370 |
| Haiku 4.5 | $0.00185 | $0.00185 |
Grade A, and why
guerillaglass 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 4d 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guerilla Glass agent guide
This file is the operational entry point for coding agents. Read it before changing code, then read the nearest nested AGENTS.md for the subsystem you touch.
Mission and current priority
Guerilla Glass is a local-first recorder and polished demo editor with a separate hosted review plane. The product workflow is Record -> Edit -> Deliver.
The current release-defining work is the Phase 2 polished demo editor core in docs/ROADMAP.md. Follow the ordered “Suggested Phase 2 PR slices”; do not let hosted delivery, Agent Mode, or broad platform expansion outrun the editor core unless the task explicitly targets those tracks.
Normative product requirements live in docs/SPEC.md. Execution status and sequencing live in docs/ROADMAP.md. Keep roadmap checkboxes synchronized with merged implementation.
Start here
| Need | Source of truth |
|---|---|
| Product requirements | docs/SPEC.md |
| Current work and PR order | docs/ROADMAP.md |
| System boundaries | docs/ARCHITECTURE.md |
| Change propagation | docs/CHANGE_MAP.md |
| Review expectations | REVIEW.md |
| Timeline semantics | docs/TIMELINE_EDITING_DESIGN.md |
| Background framing v1 | docs/BACKGROUND_FRAMING_DESIGN.md |
| Accessibility and shortcuts | docs/DESKTOP_ACCESSIBILITY.md |
| Release hardening | docs/RELEASE_HARDENING.md |
Documents named ENGINE_CONTRACT_V2_* and MIGRATION.md preserve migration history. They are useful rationale, but are not the active backlog unless docs/ROADMAP.md references them.
Repository map
apps/desktop-electrobun: Electrobun Bun host and React Creator Studio.apps/web: TanStack Start and Convex hosted review/auth shell.packages/engine-contract: Effect Schema/HttpApisource of truth and generated OpenAPI.packages/engine-client: Effect-native engine client and process launcher.packages/review-protocol: hosted review DTOs and schemas.packages/ui: shared UI primitives and styles.engines/macos-swift: production macOS capture/export engine.engines/native-foundation: shared Rust HTTP/native behavior.engines/linux-native,engines/windows-native: parity shells; do not imply production media parity.engines/protocol-rust,engines/protocol-swift: generated protocol bindings plus explicitly owned helpers/templates.messagesandproject.inlang: localization sources.Scripts: canonical generation, quality, coverage, and benchmark automation.vendor: pinned upstream source; avoid editing unless the task explicitly updates or patches a vendor.
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.
- 4d ago First seen · 127 lines · 1,849 tokens per session scan A 32208a594e3c
guerillaglass AGENTS.md is an instructions file published in the GitHub repository okikeSolutions/guerillaglass (10 stars, last pushed 27d ago), licensed MIT. It adds 1,849 tokens to every session, about $0.0092 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
EffectPatterns copilot-instructions.md
Copilot instructions for PaulJPhilp/EffectPatterns, covering effect-patterns ai agent instructions, project overview, key architectural patterns, service pattern and dependency injection.
ReignsAgent AGENTS.md
AGENTS.md instructions for Sisyphe42/ReignsAgent, covering reignsagent system specification & agent guidelines, 0. project goal, 1. core architecture (decoupled design), 2. extensibility specification and 3. maintenance protocols.
speech-studio AGENTS.md
Instructions for soniqo/speech-studio, covering agent instructions, project, stack, sibling repos under /repos/ and build.
visual-multimedia AGENTS.md
AGENTS.md instructions for CheshireMew/visual-multimedia, covering visual-multimedia 项目协作规则, 项目定位, 共享边界与唯一真源, 何时必须联动 mediaflow pro and 联动修改规则.
EffectPatterns CLAUDE.md
Claude Code instructions for PaulJPhilp/EffectPatterns: For architecture (workspace layout, MCP server, database, config, commands, debugging, important files), use.
leclap AGENTS.md
Instructions for heristop/leclap, covering agents.md, project overview, repository layout, setup and commands.