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/kevdoran/projector/claude-mdgit clone --depth 1 https://github.com/kevdoran/projectorWrote 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/kevdoran/projector/claude-md)<a href="https://agentmods.dev/instructions/kevdoran/projector/claude-md"><img src="https://agentmods.dev/badge/instructions/kevdoran/projector/claude-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 | $0.01251 | $0.01251 |
| Opus 5 | $0.00626 | $0.00626 |
| Sonnet 5 | $0.00250 | $0.00250 |
| Haiku 4.5 | $0.00125 | $0.00125 |
Grade A, and why
projector CLAUDE.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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — Coding Agent Conventions for pj
Quick Reference
# Build (preferred — injects version info from git)
make build
# Run tests (all packages)
make test
# Vet
make vet
# Tidy dependencies
make tidy
# Run a single package (no make target — use go directly)
go test -v -race -count=1 ./internal/git/...
Package Overview
| Package | Responsibility |
|---|---|
cmd/projector |
Cobra root + one file per subcommand (projects.go, list.go, create.go, desc.go, open.go, path.go, addrepo.go, archive.go, repair.go, restore.go, delete.go, version.go, config.go, config_run.go, config_list.go, config_get.go, config_set.go, config_unset.go). No business logic — delegate to internal packages. |
internal/config |
GlobalConfig struct, EditorConfig struct, Load/Save/ResolveBase/Validate. TOML I/O for ~/.config/projector/config.toml (honors $XDG_CONFIG_HOME); Load falls back to and copy-migrates the legacy ~/.projector/projector-config.toml. |
internal/project |
ProjectConfig struct, Load/Save/ListAll/FindProjectDir/ValidateName/DiscoverWorktrees. TOML I/O for <projects-dir>/<name>/.projector.toml. |
internal/git |
Thin wrappers around the git executable: RunGit, WorktreeAdd, WorktreeAddDetached, WorktreeRemove, WorktreeRepair, WorktreeList, WorktreeForBranch, StatusPorcelain, RefExists, BranchExists, BranchCheckedOut, CurrentBranch, AvailableBranchName, BranchNameFromRef, Remotes, DefaultRemote, RemoteForRef, Fetch, FetchRef, HasUnpushedCommits, HeadSHA, MinVersionCheck. |
internal/repo |
Repo struct, Discover (non-recursive scan of search dirs), ResolveRepos (name or abs-path lookup). |
internal/tui |
SelectRepos (huh multi-select), SelectEditor (huh single-select, installed editors only), EditorOption (with Terminal field), ExpandHome (tilde expansion). |
Conventions
TOML Keys
All TOML keys are kebab-case: projects-dir, repo-search-dirs, created-at, archived-at, default-base, repo-name, repo-path, worktree-path, archived-worktrees.
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 · 93 lines · 1,251 tokens per session scan A bdb0afce2d02
projector CLAUDE.md is an instructions file published in the GitHub repository kevdoran/projector (4 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 1,251 tokens to every session, about $0.0063 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
uzomuzo-oss AGENTS.md
AGENTS.md instructions for future-architect/uzomuzo-oss, covering agents.md, build & test, architecture (ddd), language policy and this repository is public.
tapd-ai-cli AGENTS.md
Instructions for studyzy/tapd-ai-cli, covering agents.md, 项目简介, 技术栈, 构建与开发命令 and 构建（版本号从 git tag 注入）.
linctl CLAUDE.md
Instructions for KyaniteHQ/linctl, covering claude.md, domain language — read before naming anything, build / test / lint (go-task), generated graphql client and live & integration tests.
jira-mcp CLAUDE.md
Instructions for nguyenvanduocit/jira-mcp, covering claude.md, project overview, development commands, build the project and run tests.
things3-cli CLAUDE.md
Claude Code instructions for ossianhempel/things3-cli, a project described as: CLI for Things 3, implemented in Go.
greatestworks gw.instructions.md
Instructions for phuhao00/greatestworks, covering greatestworks 项目 ai 协作说明（必读）, 工作流程补充要求, 项目上下文速览, 代码组织与分层约定（ddd） and 依赖与模块策略.