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/solo-agent/solo/agents-mdgit clone --depth 1 https://github.com/solo-agent/soloWhat 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 | $0.00306 | $0.00306 |
| Opus 5 | $0.00153 | $0.00153 |
| Sonnet 5 | $0.00061 | $0.00061 |
| Haiku 4.5 | $0.00031 | $0.00031 |
Grade A, and why
solo 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 3d 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.
What it actually says
Project testing rules
- Before implementation, produce a complete architecture design covering the domain model, ownership and lifecycle, data flow, persistence, APIs, frontend state, failure recovery, migration, and compatibility with existing Solo behavior.
- Review every change from the whole-project perspective. Trace the existing frontend, server, daemon, local agent runtime, session, and database paths before changing one feature path.
- All new product-flow validation must use real components and dependencies; do not use mocks.
- End-to-end tests must use the real frontend, API server, and PostgreSQL database.
- Do not mock HTTP routes, backend services, or database behavior.
- Validation must cover frontend behavior and the complete backend path, including the real local agent runtime when the feature invokes agents; UI-only tests are not sufficient.
- Report E2E success only after verifying the user-visible result and persisted database state.
Project service lifecycle rules
- An agent may start or restart the Solo frontend, API server, and daemon when requested, but only by running
make rebuildfrom the repository root. Tests must reuse that make-managed stack. - Never create
launchctljobs or manually start these services with direct binaries,go run,npm run dev,nohup, or custom background commands. - When the user asks to stop the project, use
make stopand verify that ports 3000, 8080, and 8081 are no longer listening. Do not restart services unless the user explicitly asks.
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.
- 3d ago First seen · 16 lines · 306 tokens per session scan A 0adca1eec102
solo AGENTS.md is an instructions file published in the GitHub repository solo-agent/solo (539 stars, last pushed 3d ago), licensed MIT. It adds 306 tokens to every session, about $0.0015 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
opencode-ensemble AGENTS.md
AGENTS.md instructions for hueyexe/opencode-ensemble, covering opencode-ensemble — agent guidelines, what this is, design context, users and brand personality.
opencode-hive AGENTS.md
Instructions for rretsiem/opencode-hive, covering opencode hive, directory structure, ai setup instructions, step 1: discover available models and step 2: detect target project stack.
clankerchat AGENTS.md
Instructions for simozampa/clankerchat: Preserve changes you did not create. Keep edits small and use Conventional Commits without attribution trailers. Run npm run check before declaring work complete.
clankerchat CLAUDE.md
Instructions for simozampa/clankerchat: Preserve concurrent changes, make small Conventional Commits without attribution trailers, and run npm run check before declaring work complete.
swarm-tools AGENTS.md
AGENTS.md instructions for joelhooks/swarm-tools, covering monorepo guide: bun + turborepo, critical: plugin wrapper must be self-contained, critical: no bd cli commands, critical: single global database architecture and why single database?.
lambda CLAUDE.md
Instructions for adriancarriger/lambda, covering lambda - multi-agent development, tmux windows, personality, sample app and e2e tests.