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 commands/iliaal/codesage/codesage-onboardgit clone --depth 1 https://github.com/iliaal/codesageWrote 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/commands/iliaal/codesage/codesage-onboard)<a href="https://agentmods.dev/commands/iliaal/codesage/codesage-onboard"><img src="https://agentmods.dev/badge/commands/iliaal/codesage/codesage-onboard.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 | $0.00028 | $0.01017 |
| Opus 5 | $0.00014 | $0.00508 |
| Sonnet 5 | $0.00006 | $0.00203 |
| Haiku 4.5 | $0.00003 | $0.00102 |
Grade A, and why
codesage-onboard 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Onboard a project to CodeSage
Wraps ${CLAUDE_PLUGIN_ROOT}/bin/codesage-onboard and reports results.
Unlike plugins that register one MCP server per project, CodeSage uses one global MCP (codesage) that routes every call by the absolute project path argument. The first onboard registers that MCP; every subsequent onboard on a different project reuses it.
Step 1: Validate arguments
Arguments are in $ARGUMENTS. The first positional arg must be a path that exists. If it doesn't look like a directory, stop and ask the user what to do.
Default device is gpu. The binary errors out loudly if CUDA is requested but unavailable, so don't try to be clever — just pass the user's flag through.
Step 2: Run the onboarding script
Invoke via Bash:
${CLAUDE_PLUGIN_ROOT}/bin/codesage-onboard $ARGUMENTS
The script runs seven steps: global MCP registration (idempotent), codesage init, set device in config.toml, codesage index, install git hooks (if repo), codesage git-index for V2b history (if repo), and write the agent hint. The hint lands in .claude/CLAUDE.md by default, or in a marked block inside a gitignored AGENTS.md if the repo keeps one. --refresh-hint overwrites an existing hint; --hint-only rewrites just the hint and skips MCP registration, init, indexing, and hooks.
Indexing on CUDA is fast (seconds to a few minutes for most repos). On CPU it's 5-10× slower. If the script looks like it will run longer than ~2 minutes, use run_in_background: true and poll the output file.
Step 3: Parse and report
Report what the script actually did, not what it planned:
- Whether the global
codesageMCP was newly registered or already present - Number of chunks indexed (the
Semantic:line ofcodesage indexoutput).codesage indexdoes not emit a language breakdown; runcodesage overviewor theproject_overviewMCP tool if you want one. - Whether git hooks were installed
- Whether
.claude/CLAUDE.mdwas written or already existed - Any CUDA-fallback warnings from the script output
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 · 85 lines · 28 tokens per session scan A 142a8eff89ea
codesage-onboard is a command published in the GitHub repository iliaal/codesage (20 stars, last pushed 4d ago), licensed MIT. It adds 28 tokens to every session and 1,017 once invoked, about $0.0001 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 commands, from other repositories
rb-setup
First-time setup. Configure the LLM API key, or a no-API-key local host runner (Codex / Trae / Claude / any headless CLI) that RepoBrain uses for codebase Q&A and refresh. / 首次 setup，配置 RepoBrain 代码问答与 refresh 所需的 LLM API key，或无需 API key 的本地 host runner（Codex / Trae / Claude / 任意无头 CLI）。.
rb-ask
Ask a question about the current project's codebase via the repobrain knowledge hub. / 通过 repobrain 知识库询问当前项目代码。.
rb-refresh
Rebuild the repobrain project knowledge base after significant changes. / 在重要改动后重建 repobrain 项目知识库。.
rb-init
Scaffold a new multi-agent repository from the RepoBrain template (invokes agent-repo-init skill). / 基于 RepoBrain 模板创建新的多智能体仓库。.
rp-build-cli
Build with rp-cli context builder → chat → implement.
rp-investigate-cli
Deep codebase investigation and architecture research with rp-cli commands.