The Compose setup publishes the API on 127.0.0.1 by default and the API has no login of its own unless you set an API key. The settings below control network binding, database password, API-key auth and webhook/approval hardening. They come straight from .env.example.
Docker Compose security defaults
Postgres password for the db, app, voice and worker containers. It only applies when the database volume is first created; to change an existing database also run ALTER USER ragleap WITH PASSWORD '...'. Use letters and digits only (it goes into a URL).
# POSTGRES_PASSWORD=
The API is published on 127.0.0.1 by default. Use 0.0.0.0 only behind a firewall or reverse proxy that adds authentication: RAGLEAP_API_KEY is its only protection (no lockout or rate limiting yet).
# APP_BIND=127.0.0.1
Host ports (the number on your computer, not inside Docker). Defaults are shown. The installer picks the next free port by itself when a default is already used on your computer.
# APP_PORT=8000
# DB_PORT=5433
# VOICE_PORT=8766
# NEO4J_HTTP_PORT=7475
# NEO4J_BOLT_PORT=7688
# REDIS_PORT=6380
Webhook and approval hardening
Telegram: set a secret and register the SAME value as secret_token when calling setWebhook. Without it, webhook requests are now rejected (403).
# TELEGRAM_WEBHOOK_SECRET=
WhatsApp (Twilio): requests without a valid X-Twilio-Signature are rejected (needs TWILIO_AUTH_TOKEN). Local testing only - accept unsigned webhook requests (logs a warning each time):
# TELEGRAM_ALLOW_UNSIGNED=true
# WHATSAPP_ALLOW_UNSIGNED=true
WhatsApp (Meta): WHATSAPP_PROVIDER=meta sends and receives via Meta's WhatsApp Cloud API. META_ACCESS_TOKEN and META_PHONE_NUMBER_ID come from the Meta developer portal (WhatsApp > API Setup), META_VERIFY_TOKEN is any string you also enter as the webhook verify token there, and META_APP_SECRET is the app secret used to check each webhook request's X-Hub-Signature-256 (requests without a valid signature are rejected, 403).
# WHATSAPP_PROVIDER=meta
# META_ACCESS_TOKEN=
# META_PHONE_NUMBER_ID=
# META_VERIFY_TOKEN=
# META_APP_SECRET=
YES/NO approvals are accepted ONLY from the configured owner: the sender must match approval_channel + approval_target set via POST /autonomy.
Optional API-key auth for core/api.py
Unset by default: the API has NO authentication (fine for 127.0.0.1-only access, dangerous if this port is ever reachable beyond localhost). When set, every endpoint except /health and /webhook/* (which verify platform signatures themselves) requires header: X-API-Key: <this value>
# RAGLEAP_API_KEY=
Treat answers built from your documents as untrusted content (default on). When on, an action that follows such an answer needs owner approval even in full autonomy mode, because an ingested document (an email attachment, a synced folder) can carry hidden instructions. Set to false only if you control every document you ingest.
# TAINT_RETRIEVED_DOCUMENTS=true
How long to keep the audit trail, in days. Empty or 0 keeps everything (the default); otherwise at least 7. Deleting history removes evidence: ship the AUDIT log lines to your own log store first, and see docs/audit-retention.md. Finished agent runs only; a run that is running or waiting for approval is never deleted.
# AUDIT_RETENTION_DAYS= # who changed what (audit_events)
# ACTION_LOG_RETENTION_DAYS= # what agents did and who approved it (autonomy_log)
# AGENT_RUN_RETENTION_DAYS= # finished agent runs and their steps (agent_runs)
Put the request id (the X-Request-ID of the call) in every app log line, like "INFO:core.api:[3f9c...] message". The AUDIT lines always carry it. Off by default so the log format stays as it was. See docs/audit-log-lines.md.
# LOG_REQUEST_ID=falseSecurity policy
Supported Versions
Security fixes are made against the latest published release only. There is no long-term-support branch at this stage of the project.
| Component | Supported | Notes |
|---|---|---|
ragleap-core (this repository, GitHub releases and the ragleap-app image) |
Latest release (currently 0.13.x) | Older releases: please upgrade |
Python packages in packages/ (for example ragleap-rag, ragleap-security) |
Latest PyPI release of each package | Older releases: please upgrade |
Published advisories are listed at
github.com/ragleap/ragleap-core/security/advisories.
For example, GHSA-fg9g-772q-rr9c (voice WebSocket server had no
authentication) affected ragleap-core v0.1.0 to v0.12.0 and is fixed in
v0.12.1.
Reporting a Vulnerability
Please do not open a public GitHub issue for security vulnerabilities.
Instead, report it privately using one of these methods:
- GitHub Security Advisories (preferred): open a private report via github.com/ragleap/ragleap-core/security/advisories/new
- Email: send details to the address listed on the maintainer's GitHub profile (@antonyrag).
Please include:
- A description of the vulnerability and its potential impact
- Steps to reproduce, or a minimal proof-of-concept
- The affected component and version(s)
What to Expect
This is a small, volunteer-based open-source project. There is no dedicated security team and no formal SLA. Here is the realistic process:
- You will get an acknowledgment as soon as a maintainer sees the report (typically within a few days).
- If confirmed, a fix is prioritized over other roadmap work and shipped as a patch release. Confirmed issues are published as a GitHub Security Advisory with the affected and fixed versions.
- Credit is given to the reporter in the advisory and release notes, unless you prefer to stay anonymous: just say so in your report.
- Low-severity issues (for example, requiring an already-compromised API key, or affecting only local or development use with no realistic production exposure) may be fixed on a slower, non-emergency timeline.
Scope
This policy covers the code in this repository, including ragleap-core and
the packages under packages/. It does not cover:
- Vulnerabilities in third-party dependencies (please report those upstream; flagging them here is still welcome so they can be tracked and patched on our side too)
- Deployments run by other people: the configuration, hosting, network exposure, keys and logs of an installation are the operator's responsibility
- Private repositories outside this project
Known Limitations (Honest Disclosure)
In the spirit of this project's "verified claims, not marketing claims" standard:
- RagLeap has not undergone a formal third-party security audit.
- Approval-gated autonomy and the taint rule reduce risk but do not guarantee
protection against prompt injection. The Stress Office report in
docs/stress-office.mdis a small test with simple injections, not a security certificate. - The
ragleap-ragguardrail hooks (input_guardrailsandoutput_guardrails) are extension points for user-supplied validators. The library does not itself guarantee protection against prompt injection, PII leakage, or malicious document content unless you configure guardrails for your use case. - Container images and packages are not yet signed, and a software bill of
materials is published only for
ragleap-rag. - RagLeap makes no compliance claim (for example CERT-In, SOC 2 or ISO 27001). Operators who must meet such requirements should assess their own deployment.