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 skills/sopaco/cowork-forge/codegraph-skillnpx skills add sopaco/cowork-forge --skill codegraph-skillgit clone --depth 1 https://github.com/sopaco/cowork-forgeWhat 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 | $0.00032 | $0.01164 |
| Opus 5 | $0.00016 | $0.00582 |
| Sonnet 5 | $0.00006 | $0.00233 |
| Haiku 4.5 | $0.00003 | $0.00116 |
Grade A, and why
codegraph-skill scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CodeGraph Skill
CodeGraph provides a pre-indexed AST code graph for this project.
Terrain uses CLI only — do not run codegraph install (that configures MCP/agent rules separately).
Resolve the CodeGraph command (read first)
Use conventional paths only — never machine-specific absolute paths like /Users/... or C:\Users\....
On Windows, the wrapper may be codegraph.cmd or codegraph.exe under ~/.terrain/bin/.
| Priority | Command | When |
|---|---|---|
| 1 | ~/.terrain/bin/codegraph |
Terrain env integration (see existence check below) |
| 2 | bunx codegraph |
No Terrain install; needs network once |
| 3 | npx codegraph |
Same as bunx if Bun unavailable |
Optional manifest (local, gitignored): .terrain/env/agent-tools.json — same ~/.terrain/bin/… conventions.
Existence check (cross-platform):
| Shell | Check |
|---|---|
| bash / zsh / Git Bash | [ -x ~/.terrain/bin/codegraph ] || [ -f ~/.terrain/bin/codegraph.exe ] || [ -f ~/.terrain/bin/codegraph.cmd ] |
| PowerShell | Test-Path "$HOME\.terrain\bin\codegraph*" |
In examples below, <cg> means your resolved prefix (~/.terrain/bin/codegraph or bunx codegraph).
Prerequisites
The index is per-repo under .codegraph/. If missing, initialize from the repo root:
# bash / Git Bash
if [ -x ~/.terrain/bin/codegraph ] || [ -f ~/.terrain/bin/codegraph.exe ] || [ -f ~/.terrain/bin/codegraph.cmd ]; then
~/.terrain/bin/codegraph init -i
else
bunx codegraph init -i
fi
Refresh after edits: <cg> sync
Health check:
<cg> status
<cg> status can lie ("up to date" when it is not). Observed case: index untouched for 10 days while 24 commits changed source files, status still reported fresh, and <cg> query on a symbol added days earlier returned nothing. Do not treat status as authoritative on its own.
CLI commands
| Intent | Command |
|---|---|
| Find symbol by name | <cg> query <name> |
| Who calls X | <cg> callers <symbol> |
| What X calls | <cg> callees <symbol> |
| Change blast radius | <cg> impact <symbol> |
| Tests affected by file changes | <cg> affected <files…> |
| Project file tree | <cg> files |
| Index health | <cg> status |
| Refresh after edits | <cg> sync |
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.
- 2d ago First seen · 114 lines · 32 tokens per session scan A e89680aa9b6d
codegraph-skill is a skill published in the GitHub repository sopaco/cowork-forge (90 stars, last pushed 8d ago), licensed MIT. It adds 32 tokens to every session and 1,164 once invoked, about $0.0002 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 skills, from other repositories
external-context
Invoke parallel document-specialist agents for external web searches and documentation lookup.
reflect
Review recent work, find repeated workflow patterns, and suggest reusable skills, agents, commands, config changes, or playbooks. Use when the user asks to learn from past sessions, improve recurring workflows, or identify what should be turned into reusable agent instructions.
codemap
Generate comprehensive hierarchical codemaps for UNFAMILIAR repositories. Expensive operation - only use when explicitly asked for codebase documentation or initial repository mapping.
scientific-schematics
Create publication-quality scientific diagrams using Nano Banana 2 AI with smart iterative refinement. Uses Gemini 3.1 Pro Preview for quality review. Only regenerates if quality is below threshold for your document type. Specialized in neural network architectures, system diagrams, flowcharts, biological pathways…
commit-push-pr
Commit selected local changes, push the branch, and create or update a GitHub pull request with BitFun attribution. Use when the user asks to 提交 PR、提代码、commit and push、开 PR、create a pull request, or wants a Claude Code-like one-command PR publishing flow from BitFun.
openpets
Use when the user asks to install, configure, verify, troubleshoot, or understand OpenPets; install or select a pet; connect Claude Code, OpenCode, Cursor, Codex, or MCP clients; configure a project to use a specific pet; or debug openpetsstatus, openpetsreact, or openpetssay.