Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add getaxonflow/axonflow-claude-plugin/plugin install axonflowWrote 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/getaxonflow/axonflow-claude-plugin/axonflow-login)<a href="https://agentmods.dev/commands/getaxonflow/axonflow-claude-plugin/axonflow-login"><img src="https://agentmods.dev/badge/commands/getaxonflow/axonflow-claude-plugin/axonflow-login.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 5.1 | $0.00036 | $0.00680 |
| Opus 5 | $0.00018 | $0.00340 |
| Sonnet 5 | $0.00007 | $0.00136 |
| Haiku 4.5 | $0.00004 | $0.00068 |
Grade A, and why
axonflow-login 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 7d 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
Persist an AxonFlow credential to the plugin's local config. Two modes, selected by the argument shape:
Mode A — Pro license token (default). $ARGUMENTS is an AXON-... token. Persist it so every subsequent governed tool call carries it as the X-License-Token HTTP header. The agent's plugin-claim middleware then enriches the request context with Pro-tier metadata (extended retention, higher daily quotas, etc.).
Mode B — self-hosted / Enterprise Basic-auth credential. $ARGUMENTS begins with --self-hosted followed by <org_id> <license_key>. Persist a base64 Basic-auth credential to ~/.config/axonflow/self-hosted-auth.json so BOTH the MCP headersHelper and the per-call hooks can authenticate to your self-hosted/Enterprise agent even when AXONFLOW_AUTH is not exported into every subprocess (the gap behind axonflow-claude-plugin#94). This mirrors the durable on-disk credential Community-SaaS already caches in try-registration.json. Precedence: the AXONFLOW_AUTH env var always wins; this file is the fallback. Run:
$CLAUDE_PLUGIN_ROOT/scripts/login.sh --self-hosted '<org_id>' '<license_key>'
Surface the helper's OK … / ERR … line verbatim. The credential takes effect on the next governed tool call (hooks) and on the next Claude Code session start (MCP headersHelper).
For Mode A, parse the argument as <token>. If missing, ask the user for it. Tokens issued by the AxonFlow billing webhook always start with AXON- and are at least 32 characters long. If the input doesn't look like that, refuse and ask the user to paste the token from their welcome email.
Then run the plugin's login.sh helper using the Bash tool. The helper persists the token to ~/.config/axonflow/license-token.json (mode 0600 inside a 0700 directory) and validates the AXON- prefix locally before writing.
$CLAUDE_PLUGIN_ROOT/scripts/login.sh '<the AXON- token>'
The helper writes a single-line OK … (success) or ERR … (failure) to stdout. Surface the line verbatim, then add the operator hint:
"Pro tier license saved. Restart any open Claude Code session for the new token to take effect on the MCP server connection (the plugin's
headersHelperonly fires at session start; the per-callpre-tool-checkhook picks the token up immediately on its next invocation)."
If the user prefers env-var configuration, tell them: "Alternatively, set AXONFLOW_LICENSE_TOKEN=<your AXON- token> in your shell profile — the env var takes precedence over the saved file and avoids the on-disk credential entirely."
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.
- 7d ago First seen · 33 lines · 36 tokens per session scan A bc567b72855e
axonflow-login is a command published in the GitHub repository getaxonflow/axonflow-claude-plugin (4 stars, last pushed 2d ago), licensed MIT. It adds 36 tokens to every session and 680 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 2026-08-31.
Other commands, from other repositories
ai-act-incidents
Show real-world and research-demonstrated security incidents that map to a scanner dimension, EU AI Act article, or threat category. Surfaces OWASP LLM/ASI, NIST AI RMF, and MITRE ATLAS cross-references alongside published mitigations.
ai-act-scan
Scan a codebase for EU AI Act compliance evidence and gaps. Produces a dimension-scored report with per-file findings, architecture graph, and prioritized recommendations.
ai-act-article
Show which analyzers, compliance dimensions, and current findings in this codebase map to a specific EU AI Act article.
ai-act-ask
Answer an EU AI Act question grounded in the bundled knowledge base — verbatim statute text, obligation paraphrases, and the compound-risk taxonomy. Offline and deterministic by default; cites the articles it relies on.
ai-act-settings
View or change the scanner's settings — mode (deterministic vs assisted) and autoapply. Assisted mode lets the plugin use your own Claude Code for semantic scanning, grounded Q&A, and applying fixes.
ai-act-scan-fix
Scan a codebase, then propose concrete remediation (code edits, new files, tests) for the top compliance gaps. Does NOT auto-apply — always shows the plan first.