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/kbrdn1/gwm-cli/agents-mdgit clone --depth 1 https://github.com/kbrdn1/gwm-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 | $0.02270 | $0.02270 |
| Opus 5 | $0.01135 | $0.01135 |
| Sonnet 5 | $0.00454 | $0.00454 |
| Haiku 4.5 | $0.00227 | $0.00227 |
Grade A, and why
gwm-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 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- gwm-cli CLAUDE.md — 89% identical, 11 lines differ
How it starts
The opening of the file, as written. The whole thing — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
gwm-cli — house rules for AI assistants
This file is the project-level AGENTS.md. Anything stated here OVERRIDES defaults and applies to every contribution made via an AI assistant in this repository.
🔴 Primordial rule — Test-Driven Development is mandatory
No production code lands without a failing test that pinned the behaviour down first. This is not a guideline, it is a hard merge requirement. PRs that add or change behaviour without tests are sent back, full stop.
The TDD loop (red → green → refactor)
- Red — write a failing test that captures the new behaviour or the bug you are fixing. Run it. It MUST fail for the right reason (assertion mismatch, not a compile error in unrelated code). Commit the test alone if it helps reviewers see the contract.
- Green — write the minimum production code to make the test pass. No extra branches, no speculative abstractions.
- Refactor — clean up while the tests are green. Re-run the full suite after every refactor step.
What counts as "behaviour"
Anything observable from outside the function under test:
- A new CLI subcommand, flag, or output format → end-to-end test in
tests/cli_binary.rsviaassert_cmd. - A new public function in
src/<module>.rs→ unit test intests/<module>_tests.rs. - A new bootstrap step (file copy, guard, no-symlink, command hook) →
integration test in
tests/bootstrap_tests.rsexercising it against atempfile::TempDir. - A libgit2 worktree operation → integration test in
tests/worktree_integration.rsusingtests/common::init_repo(). - A TUI state transition → state-machine test in
tests/tui_app_tests.rs(ratatui-free).
Exceptions (narrow, must be argued in the PR description)
The bar to skip a test is "the change is observably untestable from the public surface". Concretely:
- Pure formatting / typo fixes in user-facing strings → no test required if the string is incidental (a log line, a help blurb). If the string is asserted somewhere, update the assertion.
- Dependency bumps without behaviour change → CI green is the test.
- Comments-only changes → no test required.
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 · 170 lines · 2,270 tokens per session scan A 5fb16865175c
gwm-cli AGENTS.md is an instructions file published in the GitHub repository kbrdn1/gwm-cli (133 stars, last pushed 2d ago), licensed Apache-2.0. It adds 2,270 tokens to every session, about $0.0113 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
openscreen AGENTS.md
Instructions for getopenscreen/openscreen, covering agents.md, setup commands, development principles, project layout and code style.
tabularis AGENTS.md
AGENTS.md instructions for TabularisDB/tabularis, covering agents.md, directives, gitnexus — code intelligence, always do and never do.
cs CLAUDE.md
Instructions for boyter/cs, covering claude.md, project overview, build & test commands, architecture and query pipeline (pkg/search/).
hana-developer-cli-tool-example command-documentation.instructions.md
Use when creating or updating CLI command documentation in docs/02-commands/. Generates comprehensive markdown files with Mermaid diagrams showing command flow, detailed parameter tables organized by category, syntax examples, and related commands.
hana-developer-cli-tool-example cli-command-development.instructions.md
Use when creating or updating CLI command files in bin/. Enforces consistent structure, imports, exports, error handling, internationalization, and integration with the command infrastructure. Ensures commands work seamlessly with the yargs framework, prompt system, and database connections.
hana-developer-cli-tool-example sapui5-ui-development.instructions.md
Use when creating or updating SAPUI5 user interfaces in app/resources/. Enforces consistent component structure, controller patterns, view conventions, model usage, internationalization, and integration with the common UI framework. Ensures new UIs follow established patterns for reusability, maintainability, and…