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/oceanbase/powercontext/agents-mdgit clone --depth 1 https://github.com/oceanbase/powercontextWhat 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 | $0.01028 | $0.01028 |
| Opus 5 | $0.00514 | $0.00514 |
| Sonnet 5 | $0.00206 | $0.00206 |
| Haiku 4.5 | $0.00103 | $0.00103 |
Grade A, and why
powercontext 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 yesterday.
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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
powercontext is a Python package built with uv and Hatchling. Runtime code lives in src/powercontext/; keep public package exports in src/powercontext/__init__.py and place new modules beside related code. Tests live in tests/; organize focused tests around observable behavior and keep cross-component acceptance scenarios in tests/e2e/. Documentation is split by locale under docs/en/ and docs/zh/, with RFC material in each locale's rfcs/ directory. Site configuration is in zensical.toml; generated output in site/ is not source material.
Build, Test, and Development Commands
make install: runuv syncand installprekhooks.make skills-install: install the recommended agent skills pinned inskills-lock.json; requiresnpxand a Codex restart.make check: verify the lock file, run all pre-commit hooks, and runty check.make test: runpytestwith doctest support.tox: run tests and type checks across Python 3.11, 3.12, 3.13, and 3.14.make build: cleandist/and build the wheel.make docs-test: build documentation strictly.make docs: serve documentation locally with Zensical.
Agent Skills
Use an available skill only when it matches the current task, and read its SKILL.md before applying it. Recommended
skills are pinned in skills-lock.json; contributors can install them before a new Codex session with
make skills-install.
Coding Style & Naming Conventions
Target Python 3.11+. Use PEP 8 naming: modules and functions in snake_case, classes in PascalCase, constants in UPPER_SNAKE_CASE. Ruff enforces formatting and linting with a 120-character line length; run uv run prek run -a before committing if you do not use the installed hooks. Keep comments focused on non-obvious intent rather than restating code.
Generated API Contract
openapi/powercontext.yaml is the source of truth for the HTTP API contract. The Python sources in
src/powercontext/http/_generated/ are checked in; do not edit them by hand. After changing the contract, run
make api-generate and make contract-test.
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.
- yesterday First seen · 69 lines · 1,028 tokens per session scan A 4a1b833c5ebd
powercontext AGENTS.md is an instructions file published in the GitHub repository oceanbase/powercontext (923 stars, last pushed yesterday), licensed Apache-2.0. It adds 1,028 tokens to every session, about $0.0051 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
nexus AGENTS.md
Instructions for nexus-research-lab/nexus, covering agents.md, build & validation commands, commit style, l1 — 文档地图 and 状态根契约.
framework AGENTS.md
Instructions for ai-driven-dev/framework, covering agents.md, behavior, communication, action and memory management.
framework CLAUDE.md
Instructions for ai-driven-dev/framework, covering claude.md, behavior, communication, action and memory management.
nexus CLAUDE.md
Instructions for nexus-research-lab/nexus, a project described as: Multi-agent collaboration platform for persistent, proactive AI agents across rooms, workspaces, skills, and external services.
apm python.instructions.md
Python development guidelines.
toh-framework CLAUDE.md
Instructions for wasintoh/toh-framework, covering claude.md — toh framework (repo development guide), what this is, everyday commands, verification protocol and single source, transformed per ide.