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/dleerdefi/claude-code-construction/claude-mdgit clone --depth 1 https://github.com/dleerdefi/claude-code-constructionWrote 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/dleerdefi/claude-code-construction/claude-md)<a href="https://agentmods.dev/instructions/dleerdefi/claude-code-construction/claude-md"><img src="https://agentmods.dev/badge/instructions/dleerdefi/claude-code-construction/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 | $0.00456 | $0.00456 |
| Opus 5 | $0.00228 | $0.00228 |
| Sonnet 5 | $0.00091 | $0.00091 |
| Haiku 4.5 | $0.00046 | $0.00046 |
Grade A, and why
claude-code-construction CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Construction Skills — Development
This repo contains Claude Code skills for construction document management.
For Users
Import into your project's CLAUDE.md:
@.claude/skills/construction/CLAUDE.md
Or install globally:
./setup --global
For Contributors
- Production skills live in
.claude/skills/<name>/SKILL.md— edit directly - Dev/experimental skills live in
.claude/skills/_dev/<name>/(gitignored, not shipped) - Deprecated skills are renamed with
_deprecated_prefix (prevents skill matching, keeps content accessible)
Dev Workflow
bin/dev-setup # Symlink to ~/.claude/skills/construction for live testing
bin/dev-teardown # Remove symlink
Key Rules
- Each commit = one logical change
- SKILL.md files are the canonical source — edit directly
reference/data is shared across skills — accessed via${CLAUDE_SKILL_DIR}/../../reference/scripts/Python tools are shared — accessed via${CLAUDE_SKILL_DIR}/../../scripts/- All skills write findings to
.construction/agent_findings/via graph entry pattern - Never fabricate dimensions, spec requirements, or code citations
- Skills must pass eval before moving from
_dev/to production
Codex Compatibility
- Every skill directory includes
agents/openai.yamlfor OpenAI Codex agent compatibility - When creating a new skill, copy an existing
agents/openai.yamland update the name/description
Script Allowlist
- Skills must declare an exhaustive list of allowed scripts in their SKILL.md
- Skills must NOT create custom Python scripts during execution
- All scripts live in the shared
scripts/directory or per-skillscripts/subdirectories
Authoritative SOP
- The comprehensive skill architecture SOP is at
docs/CM_SKILLS_SOP.md - Covers: three-tier progressive disclosure, YAML front matter spec, 500-line limit, refactoring lifecycle, evals framework, skill categories (Builder, Extraction, Generation, Research)
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 · 50 lines · 456 tokens per session scan A 927d43fafaf1
claude-code-construction CLAUDE.md is an instructions file published in the GitHub repository dleerdefi/claude-code-construction (35 stars, last pushed 4mo ago), licensed MIT. It adds 456 tokens to every session, about $0.0023 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
PPT-Design-Skill CLAUDE.md
Claude Code instructions for sunchaokun/PPT-Design-Skill: Use the ppt-design-skill workflow for presentation design tasks. The published pptx-designer Python library creates the editable PPTX; this repository owns the design process, installation, rendering, and PNG review.
give-me-job AGENTS.md
Instructions for kyoungbinkim/give-me-job, covering give-me-job repository instructions, default workflow, skill routing, working on this repository and development loop.
agent-website-design-skills AGENTS.md
Instructions for divyanshu-iitian/agent-website-design-skills, covering agent instructions, repository goal, how to use, skill selection and constraints.
fcp-sheets CLAUDE.md
Instructions for os-tack/fcp-sheets, covering fcp-sheets, project overview, architecture, key patterns and commands.
paperless-ngx-mcp-server CLAUDE.md
Claude Code instructions for cbsmiley/paperless-ngx-mcp-server, covering claude.md, project overview, common commands, architecture and entry points.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).