fu351/Doberman-Core

docs: publish an honest OWASP-LLM / NIST-AI-RMF control-coverage matrix 2 days ago

documentation help wanted level-4

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

OpenTelemetry AuditSink 2 days ago

help wanted good first issue level-4

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Connector research memo: Cursor and Codex CLI interception surfaces 2 days ago

help wanted level-5

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Labeled detection corpus + FPR/TPR benchmark harness 2 days ago

help wanted level-5

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

calibrate_perplexity_threshold: the model-agnostic half of the perplexity seam (OT.3) 2 days ago

enhancement help wanted good first issue roadmap level-4

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

cli: scan --json silently ignores --quiet; pin the precedence with a test and a docs line 3 days ago

enhancement good first issue level-3

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

ci: add an offline checker for internal Markdown links and anchors 3 days ago

enhancement good first issue level-3

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: CLI.md's machine-readable section omits log --jsonl and the scan evidence cap 3 days ago

documentation enhancement good first issue level-1

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: docs/CLI.md documents only 10 of the CLI's 26 commands 3 days ago

documentation enhancement good first issue level-1

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

tests: assert docs/REASON_CODES.md gives every ReasonCode a real explanation 6 days ago

enhancement good first issue level-4

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

tests: assert the objective guardrail's rule-reduction loop is exhaustively raise-only 7 days ago

enhancement good first issue level-3

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

cli: add redaction-safe JSON Lines output to doberman log 7 days ago

enhancement good first issue level-4

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: catalogue every ReasonCode and what it means (docs/REASON_CODES.md) 7 days ago

documentation enhancement good first issue level-2

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

cli: add a --quiet flag to doberman scan 7 days ago

enhancement good first issue level-2

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: add a full CLI reference page (docs/CLI.md) 7 days ago

documentation enhancement good first issue level-1

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

cli: add JSON output to doberman doctor without changing health exits 7 days ago

enhancement good first issue level-3

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

cli: add deterministic JSON output to doberman scan 7 days ago

enhancement good first issue level-3

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

cli: add --json output to doberman policy-history 7 days ago

enhancement good first issue level-3

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

adapter: guard another agent harness (Aider / OpenAI Codex CLI / Cursor / generic MCP) 7 days ago

enhancement help wanted level-7

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: document the shared shape of Doberman's host adapters 7 days ago

documentation enhancement good first issue level-3

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Cryptographically signed, append-only decision log + policy ledger 7 days ago

enhancement help wanted roadmap level-10

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Passkeys / WebAuthn as a stronger local 2FA tier 7 days ago

enhancement help wanted roadmap level-10

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

F-Monitor — a warm ambient-monitoring daemon seam for the decision engine 7 days ago

enhancement help wanted roadmap level-10

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Async / non-blocking auth-challenge mechanism (interface only) 7 days ago

enhancement help wanted roadmap level-9

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Cost-observability seam — CostObserver plugin + loop-anomaly detector 7 days ago

enhancement help wanted roadmap level-9

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Adaptive red-team harness for the Unicode-smuggling / homoglyph defenses 7 days ago

enhancement help wanted roadmap level-8

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

adapter: guard another agent harness (Aider / OpenAI Codex CLI / Cursor / generic MCP) 7 days ago

enhancement help wanted level-7

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

tests: smoke-test --help for every Typer command and group 7 days ago

enhancement good first issue level-2

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: add a targeted-test selection guide for contributors 7 days ago

documentation enhancement good first issue level-1

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: add troubleshooting for PATH-shadowed Doberman installations 8 days ago

documentation enhancement good first issue

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: correct Windows virtual environment activation commands 8 days ago

documentation enhancement good first issue

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Policy-as-code — versioned policy files checked into a repo 15 days ago

enhancement good first issue roadmap

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

rule: extend CI/CD-config protection beyond GitHub Actions (GitLab, Jenkins, CircleCI, Azure Pipelines) 17 days ago

enhancement good first issue

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

tests: property-based (Hypothesis) laws for the raise-only combine() 26 days ago

help wanted good first issue

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

cli: add a 'doberman doctor' self-check command 26 days ago

enhancement good first issue

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

secrets: add Azure & GCP service-account key patterns to the built-in detector 26 days ago

enhancement good first issue

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

examples: add a tutorial custom-Guardrail plugin package 26 days ago

documentation enhancement help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

docs: add a CONTRIBUTING.md to onboard new contributors 26 days ago

documentation good first issue

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

QA: thoroughly test everything merged to main after 2026-06-20 (#18, #21, #26–#29) about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Review request: PR #47 (red-team gate ratchet + channel coverage) — ready, needs a peer to merge about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Review request: PR #54 — Windows onboarding crash fix (encode-safe CLI output) about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Review request: host-hooks stack HK.1-HK.5.1 (PRs #32->#44) — all ready, needs a peer to review + merge about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

hosthooks: PreToolUse cwd="." fallback may load the wrong strictness mode (HK.1) about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

tests(hosthooks): isolate test_interactive_global_flag_overrides_prompt — it writes to the real ~/.claude/settings.json about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

rule-lab: harden the acceptance gate — bound --eps-fpr and enforce tune/holdout disjointness about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

tests(ood): add red-team corpus entries for invisible_only and unpaired_surrogate channels about 1 month ago

help wanted

fu351/Doberman-Core

111

Doberman is an AI agent security framework for guardrails, prompt injection defense, runtime policy enforcement, tool-use permissions, agent monitoring, audit logs, LLM safety, autonomous workflow protection and secure AI deployment.

Python

Open Issues Need Help