8000 Overview · simon-inkie/inkie-the-brain · GitHub
[go: up one dir, main page]

Skip to content

Security: simon-inkie/inkie-the-brain

Security

SECURITY.md

Security policy

Supported versions

The brain is in early-stage open-source release (v0.x). Only the latest published release receives fixes. Pre-1.0 versions may be replaced rather than patched if the fix is invasive.

Reporting a vulnerability

Please do not open a public GitHub issue for security-sensitive reports. The issue tracker is for general bugs and feature requests.

Instead, email simon@inkie.ink with:

  • A clear description of the issue (one paragraph is fine)
  • Steps to reproduce, or a minimal proof-of-concept if relevant
  • Your assessment of impact (what an attacker could do)
  • Whether you'd like to be credited in the fix changelog

You can expect:

  • Acknowledgement within 72 hours (often the same day)
  • An honest read on whether the report is in scope and what the timeline looks like
  • A public advisory and changelog entry once a fix lands, with credit unless you'd rather stay anonymous

What's in scope

  • Code execution paths in the brain itself: the hook handlers, the shared memory-tools/ shell scripts, the MCP server, indexer, observer pipeline, daemon and CLI
  • Insecure handling of secrets: API keys, .env files (including the BRAIN_ENV_FILE override and the legacy fallback location), transcript content, observation content
  • Path traversal or write-zone escape in the agent silo or brain vault layout, including the <project>/.the-brain/memory_root pointer file and the AGENT_NAME silo-routing path
  • Prompt-injection vectors that compromise the observer or the memory-injection path, including anything that can get attacker-controlled text spliced into MEMORY.md between the live-block markers, or into a SessionStart injection
  • Cross-silo recall leakage: anything that lets one agent's scoped remembering query return another silo's content when it should not
  • Cost-exhaustion paths that bypass the shared embedding gate in core/embedder/gate.ts, since that gate is the only thing standing between a malformed corpus and an unbounded embedding bill
  • Command injection in the poke-agy tmux wake path, which builds and sends keystrokes to a live terminal session

What's out of scope (you'll get a friendly "not in scope" response)

  • Vulnerabilities in third-party services we depend on (Qdrant, Google Gemini API, Anthropic API). Report those upstream.
  • Issues that require an attacker to already have arbitrary code execution on the host running the brain.
  • Issues in the user's own integration code, for example their custom hook scripts or their ~/.claude/settings.json wiring.
  • Theoretical issues without a demonstrated impact.
  • The fact that observation and reflection content is sent to a third-party LLM. That is the design, it is documented, and choosing which model sees your transcripts is the operator's call.

Why this matters

The brain holds ambient context from its user's coding sessions: code snippets, decisions, partial keys mistakenly pasted into a prompt, file paths, project structure. A compromise of the brain's storage layer or its recall API is a compromise of that ambient context. It also runs inside the agent's own hook path, so a compromise there is a compromise of the agent. We take this seriously; please flag anything that looks wrong.

There aren't any published security advisories

0