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/docxology/template/claude-mdgit clone --depth 1 https://github.com/docxology/templateWrote 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/docxology/template/claude-md)<a href="https://agentmods.dev/instructions/docxology/template/claude-md"><img src="https://agentmods.dev/badge/instructions/docxology/template/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 | $0.08083 | $0.08083 |
| Opus 5 | $0.04041 | $0.04041 |
| Sonnet 5 | $0.01617 | $0.01617 |
| Haiku 4.5 | $0.00808 | $0.00808 |
Grade B, and why
template CLAUDE.md scanned grade B with 1 finding 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
sudo tlmgr install multirow cleveref doi newunicodechar How it starts
The opening of the file, as written. The whole thing — 637 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Overview
This is a research project template with a test-driven development workflow, automated PDF generation, and multi-project support. It uses a two-layer architecture separating generic infrastructure (Layer 1) from project-specific code (Layer 2), following a thin orchestrator pattern.
How this file fits with other entry points
| File | Use when you need |
|---|---|
START_HERE.md |
The Step 0-5 cold-start path: prerequisites, clone, uv sync, pre-commit, first pipeline run, verify outputs |
README.md |
First-time setup, documentation map, contributor links |
This file (CLAUDE.md) |
Copy-paste commands, CI parity, common code patterns |
AGENTS.md |
Full pipeline semantics, validation, configuration reference, troubleshooting index |
.cursorrules |
Cursor-focused agent rules (overlap with this file by design) |
.github/AGENTS.md |
Exact CI job names, coverage thresholds, branch protection hints |
Quick Reference
| Task | Command |
|---|---|
| Interactive menu | ./run.sh |
| Reproducible run matrix (project × stage; preferred over the menu for repeatable subset runs) | cp run.config.example.yaml run.config first (the CLI errors without one), then uv run python scripts/runner/run_matrix.py (reads run.config; --dry-run to preview, --fail-fast to stop on first failure; see run.config.example.yaml) |
Secure workflow via main shell (secure subcommand) |
./run.sh --secure-run |
| Full pipeline | ./run.sh --pipeline |
| Core pipeline (no LLM) | uv run python scripts/runner/execute_pipeline.py --project {name} --core-only |
| Incremental pipeline (opt-in stage skipping) | uv run python scripts/runner/execute_pipeline.py --project {name} --incremental (also python -m infrastructure.orchestration pipeline --project {name} --incremental; default off) |
| Project pipeline tests | uv run python scripts/pipeline/stage_01_test.py --project {name} |
| Full infrastructure gate | uv run python scripts/pipeline/stage_01_test.py --infra-only --infra-scope full |
| Single test | uv run pytest path/to/test.py::test_function -v |
| Install deps | uv sync (root default-groups includes public-exemplars, covering all deterministic public-project dependencies; live research/publishing groups remain opt-in) |
| Editor Python | .venv/bin/python after uv sync (see .vscode/settings.json) |
| Public CI source paths | uv run python -m infrastructure.project.public_scope source-paths |
| Ruff (CI scope) | uv run python -m infrastructure.project.public_scope source-paths | xargs uv run ruff check --fix && uv run python -m infrastructure.project.public_scope source-paths | xargs uv run ruff format |
| Mypy (CI scope) | uv run python -m infrastructure.project.public_scope source-paths | xargs uv run mypy |
| Bandit (CI / security job) | uv run bandit -c bandit.yaml -r -ll infrastructure/ scripts/ projects/ (exclusions in bandit.yaml → exclude_dirs) |
| Pre-commit (lint stage) | pre-commit run --all-files |
| Pre-push hooks | pre-commit run --hook-stage pre-push --all-files |
| Local CI reproduction (act + fallback) | ./scripts/shell/ci_local.sh (added 2026-05-20; see docs/maintenance/ci-local.md) |
| Executable bundle (opt-in Stage 14) | uv run python scripts/runner/bundle_executable.py --project {name} |
| Archive publication dry-run (opt-in Stage 15) | uv run python scripts/runner/archive_publication.py --project {name} |
| Archive publication real deposit | uv run python scripts/runner/archive_publication.py --project {name} --providers zenodo software_heritage ipfs_pinata --commit (requires credentials — see docs/maintenance/archival-targets.md) |
| Publication runbook (standalone GitHub + real Zenodo DOI + optional mirrors) | docs/guides/publication-runbook.md |
| Unified project release (GitHub + Zenodo + DOI) | uv run python scripts/publish/publish_project_release.py --project {name} --tag v1.0.0 --repo owner/repo (opt-in; see docs/guides/publication-runbook.md) |
| Reproduction bundle (single / all public exemplars) | uv run python scripts/runner/repro_bundle.py build {name} or ... build --all-public --out output/repro_bundles (verify with ... verify <manifest>) |
| Regression tests (claim-binding tier) | uv run pytest tests/regression/ -v (claim-binding tier plus a public-roster pin; collected totals in docs/_generated/COUNTS.md, rationale in docs/maintenance/regression-testing.md) |
| Repo-wide doc linter | uv run python scripts/audit/lint_docs.py |
| Exemplar drift checker | uv run python scripts/audit/check_template_drift.py (add --strict to fail on warnings, including hardcoded doc counts) |
| Re-measure exemplar coverage (slow; runs every suite) | uv run python scripts/docgen/counts.py --verify-coverage (add --write to rewrite the recorded percentages from the measurement, then --refresh-coverage-provenance --write) |
| Module line count gate | uv run python scripts/gates/module_line_count_check.py |
| CodeGraph local commands | uv run python scripts/maintenance/codegraph_local.py commands . (optional; see docs/guides/codegraph-local.md) |
| LEANN local semantic retrieval | Optional user-level companion only; see docs/guides/leann-local.md |
| Unified health CLI | uv run python -m infrastructure.core.health (optional --gates=module-line-count) |
| Release-readiness dashboard (no network) | uv run python -m infrastructure.reporting.release_readiness --out output/release_readiness.md (add --format html; aggregates version/coverage/pipeline/docs-lint/evidence-graph from local artifacts only) |
| Opt-in security scan | uv run python scripts/gates/security_scan.py (not default pipeline/CI; missing tools report skipped, not clean) |
| Deep research dispatch (opt-in, PAID, separate from any subscription — ≈$2/report OpenAI, ≈$3–7 typical Gemini (up to ~$15 on a full manuscript)) | uv sync --group deep-research (installs SDKs + python-dotenv), keys in .env, then uv run python -m infrastructure.search.deep_research providers|submit|poll|run-project — full manuscripts are packaged in full; cost model + multi-project loop recipe in infrastructure/search/deep_research/README.md; never default pipeline/CI |
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 · 637 lines · 8,083 tokens per session scan B 91bede6adfec
template CLAUDE.md is an instructions file published in the GitHub repository docxology/template (19 stars, last pushed 3d ago), licensed Apache-2.0. It adds 8,083 tokens to every session, about $0.0404 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
LynxPrompt CLAUDE.md
Instructions for GeiserX/LynxPrompt, covering claude.md - ai agent instructions for lynxprompt, 🚀 release process (critical - read carefully), understanding the release pipeline, step-by-step release process and 1. switch to develop branch.
domain-experts AGENTS.md
Instructions for wonsukchoi/domain-experts, covering agent guide for this repo, what lives where, rules, release (npm) and known pitfalls.
karpathy-interaction-guidelines AGENTS.md
Instructions for hburaktasyurek/karpathy-interaction-guidelines, covering interaction guidelines, 1. think before acting, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.
talisman-skill AGENTS.md
Instructions for jeunecraack/talisman-skill, covering talisman — agent playbook (v1), when to run, workflow (8 phases), key features and repo structure.
trellis AGENTS.md
Instructions for craigcossairt/trellis, covering agents.md , project, what i need from agents, current state - source of truth pointers and tech stack.
claude-agents-md AGENTS.md
Instructions for hexsprite/claude-agents-md, covering agents-md — agent instructions, layout, dev, conventions and branches.