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/jujitae/invara/doctorgit clone --depth 1 https://github.com/Jujitae/invaraWhat 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 | $0.00021 | $0.00569 |
| Opus 5 | $0.00010 | $0.00284 |
| Sonnet 5 | $0.00004 | $0.00114 |
| Haiku 4.5 | $0.00002 | $0.00057 |
Grade A, and why
doctor 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.
What it actually says
The INVARA MCP server is spawned as python -m invara.mcp. When it fails to
connect, the cause is almost always the machine's python, not INVARA — and on
Windows that failure can be silent. Diagnose it for the user and tell them, in
their language, exactly what to install. Do not assume Python exists; none of
the steps below require it.
Steps
-
Find what
pythonresolves to.- Windows (PowerShell):
Get-Command python | Select-Object Source - macOS/Linux:
command -v python || command -v python3
- Windows (PowerShell):
-
Interpret:
- No
pythonat all → Python is not installed. Go to step 4. - Windows: the source is under
...\Microsoft\WindowsApps\python.exe→ this is the Microsoft Store app-execution alias, not Python. Confirm with step 3. This is the known silent case (INV-011). - A real installation path → run
python --version; if it prints < 3.12, the interpreter is below INVARA's floor. Go to step 4.
- No
-
Confirm the Store alias (Windows). Run:
python --version; $LASTEXITCODEMeasured signature of the alias when Python is not installed: exit code 9009 (or 49 through an 8-bit truncating layer), stderr of exactly 7 bytes:Python, nothing on stdout. If you see this, Python is not actually installed on this machine — the alias only pretends. -
Tell the user, bilingually (English + the user's language), all three of:
- What is missing: Python 3.12+ is not installed (or the Microsoft Store alias is shadowing it).
- How to fix it: install from https://www.python.org/downloads/ (on
Windows also:
winget install Python.Python.3.12), or disable the alias under Settings > Apps > Advanced app settings > App execution aliases. During python.org installation, check "Add python.exe to PATH". - Whose fault it is not: this failure happens before any INVARA code runs; it is the environment, not an INVARA defect.
-
After the fix, ask the user to restart Claude Code so the plugin respawns the server, then verify with
/mcpthatinvarais connected.
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 · 44 lines · 21 tokens per session scan A 5d1353410109
doctor is a command published in the GitHub repository Jujitae/invara (0 stars, last pushed 2d ago), licensed Apache-2.0. It adds 21 tokens to every session and 569 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 2026-08-31.
Other commands, from other repositories
brainblast-deploy-plan
Deployment Intelligence for Anchor programs. Run before anchor deploy so you know how much SOL you need and the exact ordered transaction sequence — instead of working it out by hand.
brainblast-rico-maps
Runs a two-part token check on any Solana contract address (CA).
brainblast-firewall
Inspect a serialized Solana transaction before signing it — the AI-agent transaction firewall.
brainblast-score
A 0–100 trust score and A–F grade for any deployed Solana program: upgrade authority (renounced > DAO > multisig > single-key), verified build, audits, directory curation, and cross-cluster parity — with a transparent factor breakdown.
init-brief
Initialize a briefloop workspace using conversational onboarding.
propose-competitors
Propose competitor candidates for a workspace based on user.md context.