fivedrisk 0.5.0__tar.gz

fivedrisk-0.5.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for describing the origin of the Work and
+      reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Support. While redistributing the Work or
+      Derivative Works thereof, You may choose to offer, and charge a
+      fee for, acceptance of support, warranty, indemnity, or other
+      liability obligations and/or rights consistent with this License.
+      However, in accepting such obligations, You may act only on Your
+      own behalf and on Your sole responsibility, not on behalf of any
+      other Contributor, and only if You agree to indemnify, defend, and
+      hold each Contributor harmless for any liability incurred by, or
+      claims asserted against, such Contributor by reason of your accepting
+      any such warranty or support.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed" page as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Loren Angoni
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+   implied. See the License for the specific language governing permissions
+   and limitations under the License.

fivedrisk-0.5.0/PKG-INFO

@@ -0,0 +1,342 @@
+Metadata-Version: 2.4
+Name: fivedrisk
+Version: 0.5.0
+Summary: Per-action AI agent risk scoring and governance: 5-dimension scoring, HITL enforcement, Markov drift detection, injection scanning, and auditable decision logs.
+Author: Loren Angoni
+License: Apache-2.0
+Project-URL: Homepage, https://github.com/theDoc001/fivedrisk
+Project-URL: Repository, https://github.com/theDoc001/fivedrisk
+Project-URL: Bug Tracker, https://github.com/theDoc001/fivedrisk/issues
+Project-URL: Changelog, https://github.com/theDoc001/fivedrisk/blob/main/CHANGELOG.md
+Keywords: ai-safety,ai-governance,agent-safety,llm-safety,prompt-injection,hitl,risk-scoring,audit-log,markov-drift,owasp,langgraph,claude,anthropic
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Security
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Typing :: Typed
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: PyYAML>=6.0
+Provides-Extra: langgraph
+Requires-Dist: langgraph>=0.2.3; extra == "langgraph"
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == "dev"
+Requires-Dist: pytest-asyncio>=0.23; extra == "dev"
+Dynamic: license-file
+
+# fivedrisk — AI Agent Risk Governance Engine
+
+[![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
+[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
+[![Tests](https://github.com/theDoc001/fivedrisk/actions/workflows/tests.yml/badge.svg)](https://github.com/theDoc001/fivedrisk/actions/workflows/tests.yml)
+
+**Per-action risk scoring and governance for AI agents.** Drop one decorator on your tool functions, or wire in the Agent SDK hooks — every action is scored, gated, and logged before it runs.
+
+```python
+from fivedrisk.hooks import gate
+
+@gate(tool_name="write_to_database", autonomy_context=2)
+async def write_record(table: str, data: dict) -> None:
+    ...  # only executes if 5D scores GREEN or YELLOW
+         # ORANGE → human approval required
+         # RED    → blocked, never runs
+```
+
+---
+
+## What it does
+
+fivedrisk scores every AI agent action on **5 risk dimensions** (0–4 each):
+
+| Dimension | What it measures |
+|---|---|
+| **D** — Data Sensitivity | Public → PII → financial → credentials |
+| **T** — Tool Privilege | Read-only → write → admin → destructive |
+| **R** — Reversibility | Undoable → hard-to-undo → irreversible |
+| **E** — External Impact | Local → internal API → external → untrusted |
+| **A** — Autonomy Context | User-direct → agent-supervised → fully autonomous |
+
+5D stays deterministic. Bands signal what fivedrisk wants your stack to do; your stack chooses the LLM and the workflow.
+
+**Default 3-band:**
+
+- **GREEN** — execute, normal logging.
+- **ORANGE** — HITL approval required. fivedrisk signals; your stack handles the LLM choice. No auto model promotion.
+- **RED** — blocked.
+
+**Opt-in 4-band compliance mode** (`enable_yellow_band: true` in `policy.yaml`): surfaces YELLOW as a stable moderate-risk tier for audit queries and dashboards. Optional model escalation within YELLOW via `yellow_model_escalation: true`.
+
+fivedrisk is one layer in a defence-in-depth AI governance stack.
+
+---
+
+## Features
+
+- **5D scoring engine** — deterministic, ~40µs per action (p50) on M1, no LLM calls
+- **Markov SafetyDrift** — 16-state Markov chain detects cumulative risk across action sequences; catches compositional attacks that individual scoring misses
+- **Session accumulator** — O(1) counter-based drift tracking for the common case
+- **Injection scanner** — 24+ regex patterns covering GPT-5/Opus-era evasion (Base64, zero-width Unicode, role hijacks, encoded exec calls)
+- **Output leakage scanner** — PII, credentials, crypto keys, injection-echo detection
+- **`@gate` decorator** — wrap any sync or async function with full 5D gating
+- **Agent SDK hooks** — `fivedrisk_pre_tool` / `fivedrisk_post_tool` for Anthropic Agent SDK
+- **LangGraph node** — drop-in integration for LangGraph pipelines
+- **Destination policy** — allowlist/denylist for outbound endpoints
+- **Audit log** — append-only SQLite decision log plus optional NDJSON event stream for SIEM delivery
+- **Policy floor enforcement** — floor rules in `policy.yaml` cannot be overridden at runtime
+- **Defence-in-depth test suite** — pytest markers per OWASP LLM Top 10 category, plus 39-scenario attack-class benchmark (`python -m fivedrisk benchmark`)
+- **424 tests** with 0 failures
+
+---
+
+## Install
+
+```bash
+pip install fivedrisk
+```
+
+For LangGraph integration:
+```bash
+pip install "fivedrisk[langgraph]"
+```
+
+---
+
+## Quick start (30 seconds)
+
+```python
+from fivedrisk import classify_tool_call, score, load_policy, Band
+
+policy = load_policy("policy.yaml")  # or use defaults
+action = classify_tool_call("Bash", {"command": "rm -rf /tmp/cache"}, policy)
+result = score(action, policy)
+
+print(result.band)                      # Band.ORANGE
+print(result.rationale)  # "ORANGE — Bash: Reversibility=3 (≥ ORANGE threshold 3)"
+print(result.routing)    # RoutingDecision(model_floor=M3, approval_required=True)
+```
+
+**With the `@gate` decorator:**
+```python
+from fivedrisk.hooks import gate, configure
+from fivedrisk import load_policy
+
+configure(policy=load_policy("policy.yaml"))
+
+@gate(tool_name="send_email", autonomy_context=1)
+def send_email(to: str, body: str) -> None:
+    # only executes if 5D scores GREEN or YELLOW
+    smtp.send(to, body)
+```
+
+**With Anthropic Agent SDK:**
+```python
+from fivedrisk.hooks import fivedrisk_pre_tool, fivedrisk_post_tool
+
+# Register as PreToolUse and PostToolUse hooks in your agent
+```
+
+**With LangGraph:**
+```python
+from fivedrisk.langgraph_node import fivedrisk_gate_node
+# Add fivedrisk_gate_node to your StateGraph before any tool-executing node
+```
+
+---
+
+## SafetyDrift — why sequence risk matters
+
+A single READ of a config file scores GREEN. But 10 GREENs followed by a write to an external API using credentials extracted two steps earlier is a RED sequence. Most tools miss this.
+
+fivedrisk tracks cumulative session state via a **16-state Markov chain** over `(data_exposure_tier × activity_risk_tier)`. When absorption probability into a dangerous state crosses 0.3, the next action is escalated to ORANGE. At 0.7, it's escalated to RED.
+
+```python
+from fivedrisk.markov import MarkovDriftTracker, make_default_transition_matrix
+
+tracker = MarkovDriftTracker(make_default_transition_matrix(), session_id="abc")
+bump = tracker.record(scored_action)
+if bump:
+    print(f"Drift: {bump.reason}, escalated to {bump.escalated_band}")
+```
+
+---
+
+## Policy configuration
+
+```yaml
+# policy.yaml
+thresholds:
+  red_threshold: 4
+  orange_threshold: 3
+  orange_score: 1.8
+  yellow_score: 1.0
+
+[floor]
+# These rules block regardless of per-action score
+- tool_name: "Bash"
+  command_contains: "DROP TABLE"
+  band: RED
+  reason: "floor:no-destructive-sql"
+```
+
+---
+
+## Benchmark
+
+```bash
+python -m fivedrisk benchmark
+```
+
+Runs 39 offline attack scenarios across: prompt injection (12 categories), output leakage (8 categories), runtime tool misuse (19 scenarios). No external API calls. Deterministic. Safe to run in CI.
+
+---
+
+## Performance
+
+Measured on Apple M1, single-thread. Reproducible from `benchmarks/bench_minimal.py`.
+
+| Operation | p50 | p99 |
+|---|---|---|
+| **5D core (classify + score)** | **40µs** | **42µs** |
+| Injection scan, 30 char clean | 11µs | 12µs |
+| Injection scan, 3000 char clean | 669µs | 685µs |
+| Leakage scan, 200 char clean | 23µs | 23µs |
+| 5D + injection + leakage scan | 64µs | 65µs |
+| 5D + Markov drift | 43µs | 44µs |
+| 5D + SQLite audit-log write (I/O) | 440µs | 1ms |
+| `@gate` sync overhead (incl. log write) | 439µs | 1.1ms |
+| `@gate` async overhead | 421µs | 660µs |
+
+External optional layers (not in fivedrisk):
+
+- [PromptArmor](https://promptarmor.com): ~20-30ms typical
+- [LLM Guard](https://llm-guard.com) (token scanners): ~10-15ms typical
+- [LLM Guard](https://llm-guard.com) (ML scanners): ~50-200ms typical
+
+Injection scanner is linear in input length; for large RAG contexts, chunk and parallelize. Run the bench script on your target hardware for numbers that match your install.
+
+---
+
+## Audit log
+
+fivedrisk produces an append-only decision log entry for every agent action. Each entry records:
+- Risk band and rationale
+- Dimension scores (all 5 axes)
+- Model routing decision and approval history
+- Session drift state
+- Injection and leakage scan results
+- Optional agent identity claim (see below)
+
+### Agent identity passthrough
+
+fivedrisk accepts opaque agent identity claims through `Action.metadata["agent_identity"]`. The string flows through unchanged into the audit log for SOC/SIEM correlation. SVID, JWT, and X.509 subject strings are supported as opaque data today.
+
+```python
+action.metadata["agent_identity"] = "spiffe://example.org/agents/triage-bot"
+```
+
+Cryptographic validation, structured parsing, and identity-aware policy hooks are post-OSS scope.
+
+### Reserved metadata keys
+
+- `agent_identity` — opaque identity claim string. Do not overwrite with arbitrary values.
+
+---
+
+## Identity capture
+
+`Action.acting_identity` is a typed pass-through primitive for the principal an action is being taken on behalf of. Distinct from `agent_identity` (the AI agent's own workload identity); `acting_identity` is who authorized the action.
+
+```python
+from fivedrisk import gate, ActingIdentity, PrincipalType, AttestationSource
+
+ai = ActingIdentity(
+    principal_id="user-42",
+    principal_type=PrincipalType.USER,
+    attestation_source=AttestationSource.HTTP_HEADER,
+)
+
+# Per-call override
+fn("...", session_id="s1", _fivedrisk_acting_identity=ai)
+```
+
+Declare `identity_required: true` in `policy.yaml` to deny actions where the caller supplied no identity. The deny surfaces as `IdentityRequiredError` and emits an `identity_required_denial` NDJSON event.
+
+Identity-aware policy evaluation beyond admission, cryptographic validation, and SPIFFE/SPIRE native binding are post-OSS scope.
+
+---
+
+## Cost management primitives
+
+Per-session token budgeting with direct DENY at @gate when a reservation would exceed the session cap.
+
+```yaml
+# policy.yaml
+max_session_budget_tokens: 100000
+max_tool_call_budget_tokens: 4096
+```
+
+```python
+from fivedrisk import gate, configure
+
+configure(event_path="audit.ndjson", default_model_class="claude-sonnet-class")
+
+@gate(tool_name="summarize", estimated_input_tokens=2000)
+def summarize(text: str, session_id: str) -> str:
+    ...
+```
+
+If the projected token spend exceeds `max_session_budget_tokens`, @gate raises `BudgetExceededError` and emits a `budget_intervention` NDJSON event.
+
+Additional Operational FinOps capabilities (Tool Manifest admission layers, useful-progress monitoring, multi-agent budget envelopes, wall-clock / retry / delegation caps) are on the project roadmap.
+
+---
+
+## Planned
+
+Future capability surfaces signalled here for search and contributor expectations. No commitment dates.
+
+- SPIFFE / MCP reference example (end-to-end workload identity demo)
+- MITRE ATLAS coverage with real tests
+- NIST AI RMF mapping
+- OWASP Agentic Top 10 coverage doc
+- Regulatory crosswalks (AI Act Article 12, NIS2, DORA, ISO 42001)
+- Decision log analysis cookbook (sample SQL for common compliance queries)
+
+---
+
+## Architecture
+
+```
+fivedrisk/
+├── schema.py        # Band, Action, ScoredAction, HITLCard, ModelClass
+├── scorer.py        # score(), model routing (§12-19)
+├── classifier.py    # classify_tool_call() with policy baselines
+├── hooks.py         # @gate, Agent SDK hooks, injection/leakage scanners
+├── drift.py         # SessionAccumulator (O(1) counter-based)
+├── markov.py        # MarkovDriftTracker, Gauss-Jordan, absorption probs
+├── detectors.py     # Versioned detector corpus (2026-04-14.2)
+├── policy.py        # Policy dataclass + YAML loader
+├── router.py        # ModelRouter, EscalationSignal
+├── logger.py        # DecisionLog (SQLite, append-only)
+├── langgraph_node.py# LangGraph integration
+├── benchmarks.py    # 39-case offline benchmark harness
+└── tests/           # 424 tests
+```
+
+**Coverage**: 14/21 governance spec sections fully implemented.
+
+---
+
+## License
+
+Apache 2.0. See [LICENSE](LICENSE).
+
+Built by [Loren Angoni](https://langoni.me). Contributions welcome.
+
+> "An ambition that doesn't get executed is a hallucination."