freeboard 0.2.0__tar.gz

freeboard-0.2.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 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 reasonable and customary use in 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 Additional Liability. 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 additional liability.
+
+   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 Kaustubh Patil
+
+   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.

freeboard-0.2.0/PKG-INFO

@@ -0,0 +1,167 @@
+Metadata-Version: 2.4
+Name: freeboard
+Version: 0.2.0
+Summary: Measure how much safety envelope an AI agent deployment has left, without ever seeing its data.
+Author: Kaustubh Patil
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/kaustubhspatil/agent-credibility
+Project-URL: Repository, https://github.com/kaustubhspatil/agent-credibility
+Project-URL: Documentation, https://github.com/kaustubhspatil/agent-credibility/blob/main/docs/DEMO.md
+Project-URL: Research, https://github.com/kaustubhspatil/agent-credibility/blob/main/docs/STUDY.md
+Keywords: ai-agents,observability,risk,actuarial,credibility,insurance,audit,compliance,eu-ai-act,opentelemetry,privacy
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Financial and Insurance Industry
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Software Development :: Quality Assurance
+Classifier: Topic :: System :: Monitoring
+Classifier: Typing :: Typed
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: dev
+Requires-Dist: pytest>=8; extra == "dev"
+Dynamic: license-file
+
+# Freeboard
+
+**Measure how much safety envelope an AI agent deployment has left — without ever seeing its data.**
+
+*Freeboard* is the distance from a ship's waterline to its deck: the margin before it is swamped, and the quantity marine underwriters measure to decide whether a vessel is insurable at all. This measures the same thing for an agent deployment.
+
+Stdlib only. No dependencies.
+
+```bash
+pip install freeboard
+```
+
+```python
+from freeboard import Recorder, ToolSpec, Capability, Reversibility
+
+rec = Recorder(
+    deployment_id="acme-prod",
+    role="customer_support",
+    tools=[
+        ToolSpec("search_orders",
+                 capabilities=frozenset({Capability.PRIVATE_DATA}),
+                 reversibility=Reversibility.REVERSIBLE),
+        ToolSpec("issue_refund",
+                 capabilities=frozenset({Capability.EXTERNAL_EFFECT}),
+                 reversibility=Reversibility.IRREVERSIBLE),
+    ],
+    declared_capabilities={Capability.PRIVATE_DATA},   # never trusted
+    state_path="/var/lib/freeboard/acme-prod.json",    # survives restarts
+)
+
+with rec.episode(task_id="ticket-8821") as ep:
+    ep.action("search_orders", output_chars=412)
+    ep.action("issue_refund")
+    ep.resolve(success=True)
+```
+
+## Content cannot leak, because it cannot get in
+
+This is not redaction. Redaction is a deny-list that strips the sensitive things someone thought of, and it has to be audited and trusted.
+
+- **No parameter accepts free text.** `action()` takes a tool name, an error flag, and `output_chars` — a *length*. Tool output never enters the SDK.
+- **Identifiers are salted and hashed** per deployment. Task ids stay linkable within a deployment without the raw value travelling.
+- **Tool names never reach the wire.** They stay in-process for entropy and capability accounting.
+- **The wire format is an allow-list.** Every field is a number, a bool, a null, a fixed enum value, or a hex digest. `validate()` refuses anything else, and `to_wire()` cannot run without it.
+
+The test that proves it, `test_canary_never_reaches_the_wire`, pushes an SSN-shaped canary into the deployment id, role, task id, tool name and description, then asserts it appears nowhere in the payload — not in a value, not in a key, not in the JSON.
+
+A complete record is counts, hashes and enums, and this is all of it:
+
+```json
+{"deployment_id":"fdf3d47fa257542d","envelope_class":"C111-I0","episode_id":"a6b1be3e8c254f59",
+ "error_rate":0.0,"escalated":false,"manifest_hash":"fa89b11ed297a47d",
+ "max_reversibility":"irreversible","n_actions":6,"n_distinct_tools":3,"n_irreversible":1,
+ "n_mutating":1,"observed_capabilities":["external_effect","private_data","untrusted_input"],
+ "output_chars":4101,"pass_index":1,"repeat_rate":0.6,"resolved":true,"reward":null,
+ "role":"customer_support","seq":0,"prev_hash":"000…","entry_hash":"5da5…",
+ "task_hash":"5ddb12eaa459eb73","tool_entropy":1.251629}
+```
+
+## The task envelope is derived, not declared
+
+A declared scope is an incentive to understate — premiums fall as scope narrows. So scope is derived from the tool manifest the agent is actually wired to, and the declaration is kept only as a cross-check. Three views are tracked and none is collapsed into a verdict:
+
+| view | source | trusted |
+|---|---|---|
+| derived | the granted tool manifest | yes |
+| observed | what the agent actually exercised | yes |
+| declared | what the customer said | no |
+
+Disagreement is the signal, with understatement weighted double. `TaskEnvelope.rule_of_two_violated` flags an agent holding all three of untrusted input, private data and external effect — a complete attack chain. `class_code` keys on capabilities rather than tool names, so two deployments with the same powers fall in the same risk class.
+
+## Tamper-evident sequencing
+
+Each record commits to the one before it:
+
+```
+entry_hash(n) = sha256( prev_hash(n) || canonical_json(record without hashes) )
+```
+
+Deleting an episode leaves a sequence gap; editing one breaks its `entry_hash`; reordering breaks both. State persists across restarts, so an ephemeral container resumes its chain rather than generating a discontinuity.
+
+**What chaining alone does not prove:** a party who deletes a record and recomputes every hash after it produces a chain that verifies perfectly. Chaining is tamper-*evident* only once the head has been committed somewhere that party cannot rewrite — which is what `Checkpoint` is for. Any EU AI Act Article 12 claim rests on the checkpoints being anchored, not on the chain existing. That limitation is a passing test in the repo, not a footnote.
+
+## OpenTelemetry
+
+`to_otel_attributes()` returns `gen_ai.*` and `agent.credibility.*` span attributes for an existing tracer, validated on the way out. No OTel dependency.
+
+## Measuring, not just recording
+
+The estimator ships with the package, in the standard library:
+
+```python
+from freeboard import components, credibility_estimate, losses_from_records, read_jsonl
+
+values, ids = losses_from_records(read_jsonl("episodes.jsonl"))
+comp = components(values, ids)
+
+comp.k                     # the credibility constant, in episodes
+comp.episodes_for_z(0.5)   # exposure until own experience carries half the weight
+credibility_estimate(observed_rate, n, comp)
+```
+
+`p̂ = Z · own experience + (1 − Z) · class prior`, with `Z = n / (n + K)`. A new
+deployment prices off its class; one with history prices off itself. Nothing is
+overridden by an average — the average is only the fallback for what has not been
+observed yet.
+
+These are the standard unbiased Bühlmann–Straub estimators, checked in the test suite
+against a closed form (for a Beta(a,b)–Bernoulli class the constant is exactly `a + b`)
+*and* against the numpy implementation used for the published research, on the real
+24,100-episode corpus, so the shipped SDK and the published numbers cannot drift apart.
+
+## Somewhere to put it
+
+`JsonlSink` appends validated records; `read_jsonl` reads them back through the same
+validation; `verify` checks the chain across the round trip. JSON Lines rather than a
+JSON array so that a process killed mid-run leaves a readable file with a complete
+prefix, instead of an unparseable truncated array.
+
+## Status
+
+Early — the API may change before 1.0.
+
+There is **no transport layer** and **no certificate renderer**. Both are deliberate.
+Transport would point at a bureau service that does not exist yet, and a certificate's
+format is decided by whoever signs it — guessing means shipping a surface to deprecate
+after the first auditor conversation.
+
+Also worth knowing before you rely on it: there is no severity model, so this estimates
+frequency and not a premium; and the keyword fallback for tools that declare no
+capabilities is deliberately weak, which is why anything inferred is counted in
+`inferred_tools` rather than quietly trusted.
+
+The research behind it — what a pooled prior across agent deployments is worth, measured on ~40,000 episodes across three corpora, with the negative controls — lives in the repository.
+
+**Source, research and guides:** https://github.com/kaustubhspatil/agent-credibility
+
+Apache-2.0.

freeboard-0.2.0/README-pypi.md

@@ -0,0 +1,138 @@
+# Freeboard
+
+**Measure how much safety envelope an AI agent deployment has left — without ever seeing its data.**
+
+*Freeboard* is the distance from a ship's waterline to its deck: the margin before it is swamped, and the quantity marine underwriters measure to decide whether a vessel is insurable at all. This measures the same thing for an agent deployment.
+
+Stdlib only. No dependencies.
+
+```bash
+pip install freeboard
+```
+
+```python
+from freeboard import Recorder, ToolSpec, Capability, Reversibility
+
+rec = Recorder(
+    deployment_id="acme-prod",
+    role="customer_support",
+    tools=[
+        ToolSpec("search_orders",
+                 capabilities=frozenset({Capability.PRIVATE_DATA}),
+                 reversibility=Reversibility.REVERSIBLE),
+        ToolSpec("issue_refund",
+                 capabilities=frozenset({Capability.EXTERNAL_EFFECT}),
+                 reversibility=Reversibility.IRREVERSIBLE),
+    ],
+    declared_capabilities={Capability.PRIVATE_DATA},   # never trusted
+    state_path="/var/lib/freeboard/acme-prod.json",    # survives restarts
+)
+
+with rec.episode(task_id="ticket-8821") as ep:
+    ep.action("search_orders", output_chars=412)
+    ep.action("issue_refund")
+    ep.resolve(success=True)
+```
+
+## Content cannot leak, because it cannot get in
+
+This is not redaction. Redaction is a deny-list that strips the sensitive things someone thought of, and it has to be audited and trusted.
+
+- **No parameter accepts free text.** `action()` takes a tool name, an error flag, and `output_chars` — a *length*. Tool output never enters the SDK.
+- **Identifiers are salted and hashed** per deployment. Task ids stay linkable within a deployment without the raw value travelling.
+- **Tool names never reach the wire.** They stay in-process for entropy and capability accounting.
+- **The wire format is an allow-list.** Every field is a number, a bool, a null, a fixed enum value, or a hex digest. `validate()` refuses anything else, and `to_wire()` cannot run without it.
+
+The test that proves it, `test_canary_never_reaches_the_wire`, pushes an SSN-shaped canary into the deployment id, role, task id, tool name and description, then asserts it appears nowhere in the payload — not in a value, not in a key, not in the JSON.
+
+A complete record is counts, hashes and enums, and this is all of it:
+
+```json
+{"deployment_id":"fdf3d47fa257542d","envelope_class":"C111-I0","episode_id":"a6b1be3e8c254f59",
+ "error_rate":0.0,"escalated":false,"manifest_hash":"fa89b11ed297a47d",
+ "max_reversibility":"irreversible","n_actions":6,"n_distinct_tools":3,"n_irreversible":1,
+ "n_mutating":1,"observed_capabilities":["external_effect","private_data","untrusted_input"],
+ "output_chars":4101,"pass_index":1,"repeat_rate":0.6,"resolved":true,"reward":null,
+ "role":"customer_support","seq":0,"prev_hash":"000…","entry_hash":"5da5…",
+ "task_hash":"5ddb12eaa459eb73","tool_entropy":1.251629}
+```
+
+## The task envelope is derived, not declared
+
+A declared scope is an incentive to understate — premiums fall as scope narrows. So scope is derived from the tool manifest the agent is actually wired to, and the declaration is kept only as a cross-check. Three views are tracked and none is collapsed into a verdict:
+
+| view | source | trusted |
+|---|---|---|
+| derived | the granted tool manifest | yes |
+| observed | what the agent actually exercised | yes |
+| declared | what the customer said | no |
+
+Disagreement is the signal, with understatement weighted double. `TaskEnvelope.rule_of_two_violated` flags an agent holding all three of untrusted input, private data and external effect — a complete attack chain. `class_code` keys on capabilities rather than tool names, so two deployments with the same powers fall in the same risk class.
+
+## Tamper-evident sequencing
+
+Each record commits to the one before it:
+
+```
+entry_hash(n) = sha256( prev_hash(n) || canonical_json(record without hashes) )
+```
+
+Deleting an episode leaves a sequence gap; editing one breaks its `entry_hash`; reordering breaks both. State persists across restarts, so an ephemeral container resumes its chain rather than generating a discontinuity.
+
+**What chaining alone does not prove:** a party who deletes a record and recomputes every hash after it produces a chain that verifies perfectly. Chaining is tamper-*evident* only once the head has been committed somewhere that party cannot rewrite — which is what `Checkpoint` is for. Any EU AI Act Article 12 claim rests on the checkpoints being anchored, not on the chain existing. That limitation is a passing test in the repo, not a footnote.
+
+## OpenTelemetry
+
+`to_otel_attributes()` returns `gen_ai.*` and `agent.credibility.*` span attributes for an existing tracer, validated on the way out. No OTel dependency.
+
+## Measuring, not just recording
+
+The estimator ships with the package, in the standard library:
+
+```python
+from freeboard import components, credibility_estimate, losses_from_records, read_jsonl
+
+values, ids = losses_from_records(read_jsonl("episodes.jsonl"))
+comp = components(values, ids)
+
+comp.k                     # the credibility constant, in episodes
+comp.episodes_for_z(0.5)   # exposure until own experience carries half the weight
+credibility_estimate(observed_rate, n, comp)
+```
+
+`p̂ = Z · own experience + (1 − Z) · class prior`, with `Z = n / (n + K)`. A new
+deployment prices off its class; one with history prices off itself. Nothing is
+overridden by an average — the average is only the fallback for what has not been
+observed yet.
+
+These are the standard unbiased Bühlmann–Straub estimators, checked in the test suite
+against a closed form (for a Beta(a,b)–Bernoulli class the constant is exactly `a + b`)
+*and* against the numpy implementation used for the published research, on the real
+24,100-episode corpus, so the shipped SDK and the published numbers cannot drift apart.
+
+## Somewhere to put it
+
+`JsonlSink` appends validated records; `read_jsonl` reads them back through the same
+validation; `verify` checks the chain across the round trip. JSON Lines rather than a
+JSON array so that a process killed mid-run leaves a readable file with a complete
+prefix, instead of an unparseable truncated array.
+
+## Status
+
+Early — the API may change before 1.0.
+
+There is **no transport layer** and **no certificate renderer**. Both are deliberate.
+Transport would point at a bureau service that does not exist yet, and a certificate's
+format is decided by whoever signs it — guessing means shipping a surface to deprecate
+after the first auditor conversation.
+
+Also worth knowing before you rely on it: there is no severity model, so this estimates
+frequency and not a premium; and the keyword fallback for tools that declare no
+capabilities is deliberately weak, which is why anything inferred is counted in
+`inferred_tools` rather than quietly trusted.
+
+The research behind it — what a pooled prior across agent deployments is worth, measured on ~40,000 episodes across three corpora, with the negative controls — lives in the repository.
+
+**Source, research and guides:** https://github.com/kaustubhspatil/agent-credibility
+
+Apache-2.0.