@mohan-cao/jev-verify 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/LICENSE +201 -0
- package/README.md +99 -0
- package/dist/index.d.ts +48 -0
- package/dist/index.d.ts.map +1 -0
- package/dist/index.js +80 -0
- package/dist/index.js.map +1 -0
- package/package.json +46 -0
package/LICENSE
ADDED
|
@@ -0,0 +1,201 @@
|
|
|
1
|
+
Apache License
|
|
2
|
+
Version 2.0, January 2004
|
|
3
|
+
http://www.apache.org/licenses/
|
|
4
|
+
|
|
5
|
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
|
6
|
+
|
|
7
|
+
1. Definitions.
|
|
8
|
+
|
|
9
|
+
"License" shall mean the terms and conditions for use, reproduction,
|
|
10
|
+
and distribution as defined by Sections 1 through 9 of this document.
|
|
11
|
+
|
|
12
|
+
"Licensor" shall mean the copyright owner or entity authorized by
|
|
13
|
+
the copyright owner that is granting the License.
|
|
14
|
+
|
|
15
|
+
"Legal Entity" shall mean the union of the acting entity and all
|
|
16
|
+
other entities that control, are controlled by, or are under common
|
|
17
|
+
control with that entity. For the purposes of this definition,
|
|
18
|
+
"control" means (i) the power, direct or indirect, to cause the
|
|
19
|
+
direction or management of such entity, whether by contract or
|
|
20
|
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
|
21
|
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
|
22
|
+
|
|
23
|
+
"You" (or "Your") shall mean an individual or Legal Entity
|
|
24
|
+
exercising permissions granted by this License.
|
|
25
|
+
|
|
26
|
+
"Source" form shall mean the preferred form for making modifications,
|
|
27
|
+
including but not limited to software source code, documentation
|
|
28
|
+
source, and configuration files.
|
|
29
|
+
|
|
30
|
+
"Object" form shall mean any form resulting from mechanical
|
|
31
|
+
transformation or translation of a Source form, including but
|
|
32
|
+
not limited to compiled object code, generated documentation,
|
|
33
|
+
and conversions to other media types.
|
|
34
|
+
|
|
35
|
+
"Work" shall mean the work of authorship, whether in Source or
|
|
36
|
+
Object form, made available under the License, as indicated by a
|
|
37
|
+
copyright notice that is included in or attached to the work
|
|
38
|
+
(an example is provided in the Appendix below).
|
|
39
|
+
|
|
40
|
+
"Derivative Works" shall mean any work, whether in Source or Object
|
|
41
|
+
form, that is based on (or derived from) the Work and for which the
|
|
42
|
+
editorial revisions, annotations, elaborations, or other modifications
|
|
43
|
+
represent, as a whole, an original work of authorship. For the purposes
|
|
44
|
+
of this License, Derivative Works shall not include works that remain
|
|
45
|
+
separable from, or merely link (or bind by name) to the interfaces of,
|
|
46
|
+
the Work and Derivative Works thereof.
|
|
47
|
+
|
|
48
|
+
"Contribution" shall mean any work of authorship, including
|
|
49
|
+
the original version of the Work and any modifications or additions
|
|
50
|
+
to that Work or Derivative Works thereof, that is intentionally
|
|
51
|
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
|
52
|
+
or by an individual or Legal Entity authorized to submit on behalf of
|
|
53
|
+
the copyright owner. For the purposes of this definition, "submitted"
|
|
54
|
+
means any form of electronic, verbal, or written communication sent
|
|
55
|
+
to the Licensor or its representatives, including but not limited to
|
|
56
|
+
communication on electronic mailing lists, source code control systems,
|
|
57
|
+
and issue tracking systems that are managed by, or on behalf of, the
|
|
58
|
+
Licensor for the purpose of discussing and improving the Work, but
|
|
59
|
+
excluding communication that is conspicuously marked or otherwise
|
|
60
|
+
designated in writing by the copyright owner as "Not a Contribution."
|
|
61
|
+
|
|
62
|
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
|
63
|
+
on behalf of whom a Contribution has been received by Licensor and
|
|
64
|
+
subsequently incorporated within the Work.
|
|
65
|
+
|
|
66
|
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
|
67
|
+
this License, each Contributor hereby grants to You a perpetual,
|
|
68
|
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
|
69
|
+
copyright license to reproduce, prepare Derivative Works of,
|
|
70
|
+
publicly display, publicly perform, sublicense, and distribute the
|
|
71
|
+
Work and such Derivative Works in Source or Object form.
|
|
72
|
+
|
|
73
|
+
3. Grant of Patent License. Subject to the terms and conditions of
|
|
74
|
+
this License, each Contributor hereby grants to You a perpetual,
|
|
75
|
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
|
76
|
+
(except as stated in this section) patent license to make, have made,
|
|
77
|
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
|
78
|
+
where such license applies only to those patent claims licensable
|
|
79
|
+
by such Contributor that are necessarily infringed by their
|
|
80
|
+
Contribution(s) alone or by combination of their Contribution(s)
|
|
81
|
+
with the Work to which such Contribution(s) was submitted. If You
|
|
82
|
+
institute patent litigation against any entity (including a
|
|
83
|
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
|
84
|
+
or a Contribution incorporated within the Work constitutes direct
|
|
85
|
+
or contributory patent infringement, then any patent licenses
|
|
86
|
+
granted to You under this License for that Work shall terminate
|
|
87
|
+
as of the date such litigation is filed.
|
|
88
|
+
|
|
89
|
+
4. Redistribution. You may reproduce and distribute copies of the
|
|
90
|
+
Work or Derivative Works thereof in any medium, with or without
|
|
91
|
+
modifications, and in Source or Object form, provided that You
|
|
92
|
+
meet the following conditions:
|
|
93
|
+
|
|
94
|
+
(a) You must give any other recipients of the Work or
|
|
95
|
+
Derivative Works a copy of this License; and
|
|
96
|
+
|
|
97
|
+
(b) You must cause any modified files to carry prominent notices
|
|
98
|
+
stating that You changed the files; and
|
|
99
|
+
|
|
100
|
+
(c) You must retain, in the Source form of any Derivative Works
|
|
101
|
+
that You distribute, all copyright, patent, trademark, and
|
|
102
|
+
attribution notices from the Source form of the Work,
|
|
103
|
+
excluding those notices that do not pertain to any part of
|
|
104
|
+
the Derivative Works; and
|
|
105
|
+
|
|
106
|
+
(d) If the Work includes a "NOTICE" text file as part of its
|
|
107
|
+
distribution, then any Derivative Works that You distribute must
|
|
108
|
+
include a readable copy of the attribution notices contained
|
|
109
|
+
within such NOTICE file, excluding those notices that do not
|
|
110
|
+
pertain to any part of the Derivative Works, in at least one
|
|
111
|
+
of the following places: within a NOTICE text file distributed
|
|
112
|
+
as part of the Derivative Works; within the Source form or
|
|
113
|
+
documentation, if provided along with the Derivative Works; or,
|
|
114
|
+
within a display generated by the Derivative Works, if and
|
|
115
|
+
wherever such third-party notices normally appear. The contents
|
|
116
|
+
of the NOTICE file are for informational purposes only and
|
|
117
|
+
do not modify the License. You may add Your own attribution
|
|
118
|
+
notices within Derivative Works that You distribute, alongside
|
|
119
|
+
or as an addendum to the NOTICE text from the Work, provided
|
|
120
|
+
that such additional attribution notices cannot be construed
|
|
121
|
+
as modifying the License.
|
|
122
|
+
|
|
123
|
+
You may add Your own copyright statement to Your modifications and
|
|
124
|
+
may provide additional or different license terms and conditions
|
|
125
|
+
for use, reproduction, or distribution of Your modifications, or
|
|
126
|
+
for any such Derivative Works as a whole, provided Your use,
|
|
127
|
+
reproduction, and distribution of the Work otherwise complies with
|
|
128
|
+
the conditions stated in this License.
|
|
129
|
+
|
|
130
|
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
|
131
|
+
any Contribution intentionally submitted for inclusion in the Work
|
|
132
|
+
by You to the Licensor shall be under the terms and conditions of
|
|
133
|
+
this License, without any additional terms or conditions.
|
|
134
|
+
Notwithstanding the above, nothing herein shall supersede or modify
|
|
135
|
+
the terms of any separate license agreement you may have executed
|
|
136
|
+
with Licensor regarding such Contributions.
|
|
137
|
+
|
|
138
|
+
6. Trademarks. This License does not grant permission to use the trade
|
|
139
|
+
names, trademarks, service marks, or product names of the Licensor,
|
|
140
|
+
except as required for reasonable and customary use in describing the
|
|
141
|
+
origin of the Work and reproducing the content of the NOTICE file.
|
|
142
|
+
|
|
143
|
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
|
144
|
+
agreed to in writing, Licensor provides the Work (and each
|
|
145
|
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
|
146
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
|
147
|
+
implied, including, without limitation, any warranties or conditions
|
|
148
|
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
|
149
|
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
|
150
|
+
appropriateness of using or redistributing the Work and assume any
|
|
151
|
+
risks associated with Your exercise of permissions under this License.
|
|
152
|
+
|
|
153
|
+
8. Limitation of Liability. In no event and under no legal theory,
|
|
154
|
+
whether in tort (including negligence), contract, or otherwise,
|
|
155
|
+
unless required by applicable law (such as deliberate and grossly
|
|
156
|
+
negligent acts) or agreed to in writing, shall any Contributor be
|
|
157
|
+
liable to You for damages, including any direct, indirect, special,
|
|
158
|
+
incidental, or consequential damages of any character arising as a
|
|
159
|
+
result of this License or out of the use or inability to use the
|
|
160
|
+
Work (including but not limited to damages for loss of goodwill,
|
|
161
|
+
work stoppage, computer failure or malfunction, or any and all
|
|
162
|
+
other commercial damages or losses), even if such Contributor
|
|
163
|
+
has been advised of the possibility of such damages.
|
|
164
|
+
|
|
165
|
+
9. Accepting Warranty or Additional Liability. While redistributing
|
|
166
|
+
the Work or Derivative Works thereof, You may choose to offer,
|
|
167
|
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
|
168
|
+
or other liability obligations and/or rights consistent with this
|
|
169
|
+
License. However, in accepting such obligations, You may act only
|
|
170
|
+
on Your own behalf and on Your sole responsibility, not on behalf
|
|
171
|
+
of any other Contributor, and only if You agree to indemnify,
|
|
172
|
+
defend, and hold each Contributor harmless for any liability
|
|
173
|
+
incurred by, or claims asserted against, such Contributor by reason
|
|
174
|
+
of your accepting any such warranty or additional liability.
|
|
175
|
+
|
|
176
|
+
END OF TERMS AND CONDITIONS
|
|
177
|
+
|
|
178
|
+
APPENDIX: How to apply the Apache License to your work.
|
|
179
|
+
|
|
180
|
+
To apply the Apache License to your work, attach the following
|
|
181
|
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
|
182
|
+
replaced with your own identifying information. (Don't include
|
|
183
|
+
the brackets!) The text should be enclosed in the appropriate
|
|
184
|
+
comment syntax for the file format. We also recommend that a
|
|
185
|
+
file or class name and description of purpose be included on the
|
|
186
|
+
same "printed page" as the copyright notice for easier
|
|
187
|
+
identification within third-party archives.
|
|
188
|
+
|
|
189
|
+
Copyright [yyyy] [name of copyright owner]
|
|
190
|
+
|
|
191
|
+
Licensed under the Apache License, Version 2.0 (the "License");
|
|
192
|
+
you may not use this file except in compliance with the License.
|
|
193
|
+
You may obtain a copy of the License at
|
|
194
|
+
|
|
195
|
+
http://www.apache.org/licenses/LICENSE-2.0
|
|
196
|
+
|
|
197
|
+
Unless required by applicable law or agreed to in writing, software
|
|
198
|
+
distributed under the License is distributed on an "AS IS" BASIS,
|
|
199
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
200
|
+
See the License for the specific language governing permissions and
|
|
201
|
+
limitations under the License.
|
package/README.md
ADDED
|
@@ -0,0 +1,99 @@
|
|
|
1
|
+
# @mohan-cao/jev-verify
|
|
2
|
+
|
|
3
|
+
The verification signal, extracted from `@mohan-cao/jev-classifier`: three Jev
|
|
4
|
+
questions about a completed exchange, collapsed into one flag.
|
|
5
|
+
|
|
6
|
+
```bash
|
|
7
|
+
npm install @mohan-cao/jev-verify
|
|
8
|
+
```
|
|
9
|
+
|
|
10
|
+
```ts
|
|
11
|
+
import { verifyResponse, formatVerifyStatus } from "@mohan-cao/jev-verify";
|
|
12
|
+
|
|
13
|
+
const result = await verifyResponse(request, response, apiKey, config);
|
|
14
|
+
// { flag: "ok", answersQuestion: 0.86, evasive: 0.08, obligationUnmet: 0.94 }
|
|
15
|
+
```
|
|
16
|
+
|
|
17
|
+
| flag | meaning |
|
|
18
|
+
| --- | --- |
|
|
19
|
+
| `ok` | nothing to report |
|
|
20
|
+
| `🤷 evasive` | declined to commit, or did not address the request |
|
|
21
|
+
| `🚩 unmet` | did not do what the request needed |
|
|
22
|
+
|
|
23
|
+
## Read this before trusting the flag
|
|
24
|
+
|
|
25
|
+
**It is too permissive, and the reason is structural.** It has produced a real
|
|
26
|
+
false negative: across a session containing demonstrably unverified assertions —
|
|
27
|
+
a claim about an npm configuration asserted without reading it, and an
|
|
28
|
+
"incompatible with monorepo tooling" verdict drawn from one issue title that was
|
|
29
|
+
later closed as a misconfiguration — **every turn reported `ok`.**
|
|
30
|
+
|
|
31
|
+
Two separate causes, and only one of them is about thresholds.
|
|
32
|
+
|
|
33
|
+
**The questions ask about the wrong property.** `VERIFY_OBLIGATION_QUESTION`
|
|
34
|
+
asks about *the request's* premise — "if the request rested on a false or
|
|
35
|
+
oversimplified premise, did the response correct it" — and about inputs the
|
|
36
|
+
request omitted. The failure above is a third thing: **the response asserted what
|
|
37
|
+
it had not established.** That is not evasion (the response committed hard), not
|
|
38
|
+
off-topic (it answered), and not an uncorrected user premise (the premise was
|
|
39
|
+
fine). Nothing here asks the direct question: *does this response assert things it
|
|
40
|
+
did not check?*
|
|
41
|
+
|
|
42
|
+
**And the input could not answer that question anyway.** The signal is handed the
|
|
43
|
+
exchange as **text only**: tool calls are filtered out before the classifier sees
|
|
44
|
+
anything, and the response is truncated. A grounding question is unanswerable
|
|
45
|
+
from that input, because the evidence of grounding — what the agent actually read
|
|
46
|
+
or ran — has already been discarded.
|
|
47
|
+
|
|
48
|
+
So the current lens is not a mistake; it is the only thing judgeable from a
|
|
49
|
+
text-only input. **The fix is a different input, not a different question**, and
|
|
50
|
+
it does not belong in this package.
|
|
51
|
+
|
|
52
|
+
### The property has a name
|
|
53
|
+
|
|
54
|
+
Separate two things that are easy to conflate:
|
|
55
|
+
|
|
56
|
+
- **Truth** — *is this claim correct?* Needs the world. Not judgeable from text.
|
|
57
|
+
- **Provenance** — *does this claim trace to something that establishes it?*
|
|
58
|
+
Judgeable from a transcript, because the transcript contains the agent's own
|
|
59
|
+
actions.
|
|
60
|
+
|
|
61
|
+
The second is studied under a name: the **provenance gap** — tool-using agents
|
|
62
|
+
"rarely specify which tool observation supports each generated claim" — plus
|
|
63
|
+
claim-centric trajectory auditing ("tracking what the agent comes to believe,
|
|
64
|
+
whether those claims are supported") and evidence tracing generally ("which
|
|
65
|
+
evidence supported each claim, whether tool calls were justified").
|
|
66
|
+
|
|
67
|
+
The uncertainty family is the **wrong** tool for this. Semantic entropy measures
|
|
68
|
+
*the model's* uncertainty, so it catches confabulation — unstable answers — not
|
|
69
|
+
stable but unfounded ones. The failure above was stated with total confidence, so
|
|
70
|
+
low entropy would have marked it reliable.
|
|
71
|
+
|
|
72
|
+
### On the word "evasive"
|
|
73
|
+
|
|
74
|
+
`evasive` is a slight misnomer for the ungrounded case, but it is kept: renaming a
|
|
75
|
+
flag changes what every consumer and every stored record means, and the label is
|
|
76
|
+
defensible for the cases it *does* catch. Treat `evasive` as "committed without
|
|
77
|
+
doing the work", not strictly as "refused to commit".
|
|
78
|
+
|
|
79
|
+
## API
|
|
80
|
+
|
|
81
|
+
| export | purpose |
|
|
82
|
+
| --- | --- |
|
|
83
|
+
| `verifyResponse(request, response, apiKey, config, signal?)` | one Jev call → `VerifyResult` |
|
|
84
|
+
| `VERIFY_ANSWER_QUESTION`, `VERIFY_EVASIVE_QUESTION`, `VERIFY_OBLIGATION_QUESTION` | the question specs |
|
|
85
|
+
| `formatVerifyStatus(result, mode)` | status string, or `undefined` to clear |
|
|
86
|
+
| `VerifyConfig` | transport settings plus `evasiveThreshold`, `answersThreshold`, `obligationThreshold`, `enabled` |
|
|
87
|
+
|
|
88
|
+
## Notes
|
|
89
|
+
|
|
90
|
+
- **Signal only.** It never retries and never rewrites the answer. Partial
|
|
91
|
+
(`aborted`/`error`) answers should not be verified at all.
|
|
92
|
+
- **Its own call.** Not shared with the phase or trajectory judgments; questions
|
|
93
|
+
sharing a call can perturb each other.
|
|
94
|
+
- **Ordering is deliberate.** Evasion is checked before obligation, so a response
|
|
95
|
+
that dodged the question is reported as evading rather than as unmet.
|
|
96
|
+
|
|
97
|
+
## License
|
|
98
|
+
|
|
99
|
+
Apache-2.0.
|
package/dist/index.d.ts
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* The verification signal.
|
|
3
|
+
*
|
|
4
|
+
* Judges a completed exchange on three questions — did the response address the
|
|
5
|
+
* request, did it avoid committing to a position the request called for, and did
|
|
6
|
+
* it meet its obligation — and reports one flag. It is a **signal only**: it does
|
|
7
|
+
* not retry or rewrite the answer.
|
|
8
|
+
*
|
|
9
|
+
* Read the README before trusting the flag. It is known to be too permissive,
|
|
10
|
+
* and the reason is structural rather than a matter of tuning.
|
|
11
|
+
*/
|
|
12
|
+
import { type FooterMode, type JevConfig, type NoulQuestionSpec, type ScoreQuestionSpec } from "@mohan-cao/jev-classifier";
|
|
13
|
+
export declare const VERIFY_ANSWER_QUESTION: NoulQuestionSpec;
|
|
14
|
+
export declare const VERIFY_EVASIVE_QUESTION: NoulQuestionSpec;
|
|
15
|
+
export declare const VERIFY_OBLIGATION_QUESTION: ScoreQuestionSpec;
|
|
16
|
+
export type VerifyFlag = "ok" | "evasive" | "unmet";
|
|
17
|
+
export interface VerifyResult {
|
|
18
|
+
flag: VerifyFlag;
|
|
19
|
+
/** P(the response addresses the actual request). */
|
|
20
|
+
answersQuestion: number;
|
|
21
|
+
/** P(the response avoids committing to a position the request calls for). */
|
|
22
|
+
evasive: number;
|
|
23
|
+
/** Expected obligation failure, 0-3. Higher means the answer did less of what was needed. */
|
|
24
|
+
obligationUnmet: number;
|
|
25
|
+
model?: string;
|
|
26
|
+
}
|
|
27
|
+
/** Transport settings plus this component's thresholds. */
|
|
28
|
+
export interface VerifyConfig extends JevConfig {
|
|
29
|
+
/** Whether verification runs by default. */
|
|
30
|
+
enabled: boolean;
|
|
31
|
+
/** P(evasive) at or above which the response is flagged as evading. */
|
|
32
|
+
evasiveThreshold: number;
|
|
33
|
+
/** P(answers the question) at or below which the response is flagged. */
|
|
34
|
+
answersThreshold: number;
|
|
35
|
+
/** Expected obligation failure (0-3) at or above which the answer is flagged. */
|
|
36
|
+
obligationThreshold: number;
|
|
37
|
+
}
|
|
38
|
+
/**
|
|
39
|
+
* Post-generation verification only. This deliberately does not retry or
|
|
40
|
+
* rewrite the answer; it produces a signal the caller can surface to the user.
|
|
41
|
+
*/
|
|
42
|
+
export declare function verifyResponse(request: string, response: string, apiKey: string, config: VerifyConfig, parentSignal?: AbortSignal): Promise<VerifyResult>;
|
|
43
|
+
/**
|
|
44
|
+
* `undefined` clears the status indicator. `off` renders nothing; `icons`
|
|
45
|
+
* renders the glyph alone; `compact` (the default) adds the label.
|
|
46
|
+
*/
|
|
47
|
+
export declare function formatVerifyStatus(result: VerifyResult, mode?: FooterMode): string | undefined;
|
|
48
|
+
//# sourceMappingURL=index.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,EAIL,KAAK,UAAU,EACf,KAAK,SAAS,EACd,KAAK,gBAAgB,EACrB,KAAK,iBAAiB,EACvB,MAAM,2BAA2B,CAAC;AAEnC,eAAO,MAAM,sBAAsB,EAAE,gBASpC,CAAC;AAEF,eAAO,MAAM,uBAAuB,EAAE,gBASrC,CAAC;AAEF,eAAO,MAAM,0BAA0B,EAAE,iBAUxC,CAAC;AAMF,MAAM,MAAM,UAAU,GAAG,IAAI,GAAG,SAAS,GAAG,OAAO,CAAC;AAEpD,MAAM,WAAW,YAAY;IAC3B,IAAI,EAAE,UAAU,CAAC;IACjB,oDAAoD;IACpD,eAAe,EAAE,MAAM,CAAC;IACxB,6EAA6E;IAC7E,OAAO,EAAE,MAAM,CAAC;IAChB,6FAA6F;IAC7F,eAAe,EAAE,MAAM,CAAC;IACxB,KAAK,CAAC,EAAE,MAAM,CAAC;CAChB;AAED,2DAA2D;AAC3D,MAAM,WAAW,YAAa,SAAQ,SAAS;IAC7C,4CAA4C;IAC5C,OAAO,EAAE,OAAO,CAAC;IACjB,uEAAuE;IACvE,gBAAgB,EAAE,MAAM,CAAC;IACzB,yEAAyE;IACzE,gBAAgB,EAAE,MAAM,CAAC;IACzB,iFAAiF;IACjF,mBAAmB,EAAE,MAAM,CAAC;CAC7B;AAED;;;GAGG;AACH,wBAAsB,cAAc,CAClC,OAAO,EAAE,MAAM,EACf,QAAQ,EAAE,MAAM,EAChB,MAAM,EAAE,MAAM,EACd,MAAM,EAAE,YAAY,EACpB,YAAY,CAAC,EAAE,WAAW,GACzB,OAAO,CAAC,YAAY,CAAC,CA+BvB;AAED;;;GAGG;AACH,wBAAgB,kBAAkB,CAChC,MAAM,EAAE,YAAY,EACpB,IAAI,GAAE,UAAsB,GAC3B,MAAM,GAAG,SAAS,CAKpB"}
|
package/dist/index.js
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* The verification signal.
|
|
3
|
+
*
|
|
4
|
+
* Judges a completed exchange on three questions — did the response address the
|
|
5
|
+
* request, did it avoid committing to a position the request called for, and did
|
|
6
|
+
* it meet its obligation — and reports one flag. It is a **signal only**: it does
|
|
7
|
+
* not retry or rewrite the answer.
|
|
8
|
+
*
|
|
9
|
+
* Read the README before trusting the flag. It is known to be too permissive,
|
|
10
|
+
* and the reason is structural rather than a matter of tuning.
|
|
11
|
+
*/
|
|
12
|
+
import { callSystemOne, parseNoulAnswer, parseScoreAnswer, } from "@mohan-cao/jev-classifier";
|
|
13
|
+
export const VERIFY_ANSWER_QUESTION = {
|
|
14
|
+
type: "noul",
|
|
15
|
+
instructions: "Does the response address the user's actual request? Addressing includes naming what the answer depends on, or asking for information it depends on; it does not require a single final verdict.",
|
|
16
|
+
criteria: {
|
|
17
|
+
true: "The response engages the actual question and gives the user something actionable or conclusive, including a clear statement of what the answer depends on.",
|
|
18
|
+
false: "The response deflects, answers a different question, or leaves the user's request unaddressed.",
|
|
19
|
+
},
|
|
20
|
+
};
|
|
21
|
+
export const VERIFY_EVASIVE_QUESTION = {
|
|
22
|
+
type: "noul",
|
|
23
|
+
instructions: "Does the response avoid committing to a position that the request calls for? Distinguish genuine evasion from a conditional answer that names the determining factors and then commits within them.",
|
|
24
|
+
criteria: {
|
|
25
|
+
true: "The response declines to take any position, or offers only generic caveats and filler, even though the request called for a conclusion.",
|
|
26
|
+
false: "The response commits to a position, or conditions its answer on explicitly named factors and commits within each. Conditional-but-committed is not evasion.",
|
|
27
|
+
},
|
|
28
|
+
};
|
|
29
|
+
export const VERIFY_OBLIGATION_QUESTION = {
|
|
30
|
+
type: "score",
|
|
31
|
+
instructions: "Given what the request did and did not provide, did the response meet its obligation? If the request rested on a false or oversimplified premise, did the response correct it? If the request omitted information needed to answer well, did the response surface those inputs (by asking, or by branching on the determining conditions) rather than giving a generic answer that would be true regardless?",
|
|
32
|
+
criteria: [
|
|
33
|
+
"Fully met. Answered as posed and addressed any false premise or missing input.",
|
|
34
|
+
"Mostly met. Minor omissions that don't change whether the answer is useful.",
|
|
35
|
+
"Partially met. Answered the literal question but left a material premise or missing input unaddressed.",
|
|
36
|
+
"Not met. Technically true but misleading or unusable, because it ignored a false premise or the information the answer actually depends on.",
|
|
37
|
+
],
|
|
38
|
+
};
|
|
39
|
+
const ANSWER_ID = "answers_question";
|
|
40
|
+
const EVASIVE_ID = "evasive";
|
|
41
|
+
const OBLIGATION_ID = "obligation_unmet";
|
|
42
|
+
/**
|
|
43
|
+
* Post-generation verification only. This deliberately does not retry or
|
|
44
|
+
* rewrite the answer; it produces a signal the caller can surface to the user.
|
|
45
|
+
*/
|
|
46
|
+
export async function verifyResponse(request, response, apiKey, config, parentSignal) {
|
|
47
|
+
const payload = await callSystemOne({ user_request: request, assistant_response: response }, {
|
|
48
|
+
[ANSWER_ID]: VERIFY_ANSWER_QUESTION,
|
|
49
|
+
[EVASIVE_ID]: VERIFY_EVASIVE_QUESTION,
|
|
50
|
+
[OBLIGATION_ID]: VERIFY_OBLIGATION_QUESTION,
|
|
51
|
+
}, apiKey, config, parentSignal);
|
|
52
|
+
const answersQuestion = parseNoulAnswer(payload, ANSWER_ID);
|
|
53
|
+
const evasive = parseNoulAnswer(payload, EVASIVE_ID);
|
|
54
|
+
const obligationUnmet = parseScoreAnswer(payload, OBLIGATION_ID).score;
|
|
55
|
+
let flag = "ok";
|
|
56
|
+
if (evasive >= config.evasiveThreshold || answersQuestion <= config.answersThreshold) {
|
|
57
|
+
flag = "evasive";
|
|
58
|
+
}
|
|
59
|
+
else if (obligationUnmet >= config.obligationThreshold) {
|
|
60
|
+
flag = "unmet";
|
|
61
|
+
}
|
|
62
|
+
return {
|
|
63
|
+
flag,
|
|
64
|
+
answersQuestion,
|
|
65
|
+
evasive,
|
|
66
|
+
obligationUnmet,
|
|
67
|
+
...(payload.model ? { model: payload.model } : {}),
|
|
68
|
+
};
|
|
69
|
+
}
|
|
70
|
+
/**
|
|
71
|
+
* `undefined` clears the status indicator. `off` renders nothing; `icons`
|
|
72
|
+
* renders the glyph alone; `compact` (the default) adds the label.
|
|
73
|
+
*/
|
|
74
|
+
export function formatVerifyStatus(result, mode = "compact") {
|
|
75
|
+
if (mode === "off" || result.flag === "ok")
|
|
76
|
+
return undefined;
|
|
77
|
+
const glyph = result.flag === "evasive" ? "🤷" : "🚩";
|
|
78
|
+
return mode === "icons" ? glyph : `${glyph} ${result.flag}`;
|
|
79
|
+
}
|
|
80
|
+
//# sourceMappingURL=index.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,EACL,aAAa,EACb,eAAe,EACf,gBAAgB,GAKjB,MAAM,2BAA2B,CAAC;AAEnC,MAAM,CAAC,MAAM,sBAAsB,GAAqB;IACtD,IAAI,EAAE,MAAM;IACZ,YAAY,EACV,kMAAkM;IACpM,QAAQ,EAAE;QACR,IAAI,EAAE,4JAA4J;QAClK,KAAK,EACH,gGAAgG;KACnG;CACF,CAAC;AAEF,MAAM,CAAC,MAAM,uBAAuB,GAAqB;IACvD,IAAI,EAAE,MAAM;IACZ,YAAY,EACV,qMAAqM;IACvM,QAAQ,EAAE;QACR,IAAI,EAAE,yIAAyI;QAC/I,KAAK,EACH,6JAA6J;KAChK;CACF,CAAC;AAEF,MAAM,CAAC,MAAM,0BAA0B,GAAsB;IAC3D,IAAI,EAAE,OAAO;IACb,YAAY,EACV,8YAA8Y;IAChZ,QAAQ,EAAE;QACR,gFAAgF;QAChF,6EAA6E;QAC7E,wGAAwG;QACxG,6IAA6I;KAC9I;CACF,CAAC;AAEF,MAAM,SAAS,GAAG,kBAAkB,CAAC;AACrC,MAAM,UAAU,GAAG,SAAS,CAAC;AAC7B,MAAM,aAAa,GAAG,kBAAkB,CAAC;AA2BzC;;;GAGG;AACH,MAAM,CAAC,KAAK,UAAU,cAAc,CAClC,OAAe,EACf,QAAgB,EAChB,MAAc,EACd,MAAoB,EACpB,YAA0B;IAE1B,MAAM,OAAO,GAAG,MAAM,aAAa,CACjC,EAAE,YAAY,EAAE,OAAO,EAAE,kBAAkB,EAAE,QAAQ,EAAE,EACvD;QACE,CAAC,SAAS,CAAC,EAAE,sBAAsB;QACnC,CAAC,UAAU,CAAC,EAAE,uBAAuB;QACrC,CAAC,aAAa,CAAC,EAAE,0BAA0B;KAC5C,EACD,MAAM,EACN,MAAM,EACN,YAAY,CACb,CAAC;IAEF,MAAM,eAAe,GAAG,eAAe,CAAC,OAAO,EAAE,SAAS,CAAC,CAAC;IAC5D,MAAM,OAAO,GAAG,eAAe,CAAC,OAAO,EAAE,UAAU,CAAC,CAAC;IACrD,MAAM,eAAe,GAAG,gBAAgB,CAAC,OAAO,EAAE,aAAa,CAAC,CAAC,KAAK,CAAC;IAEvE,IAAI,IAAI,GAAe,IAAI,CAAC;IAC5B,IAAI,OAAO,IAAI,MAAM,CAAC,gBAAgB,IAAI,eAAe,IAAI,MAAM,CAAC,gBAAgB,EAAE,CAAC;QACrF,IAAI,GAAG,SAAS,CAAC;IACnB,CAAC;SAAM,IAAI,eAAe,IAAI,MAAM,CAAC,mBAAmB,EAAE,CAAC;QACzD,IAAI,GAAG,OAAO,CAAC;IACjB,CAAC;IAED,OAAO;QACL,IAAI;QACJ,eAAe;QACf,OAAO;QACP,eAAe;QACf,GAAG,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,KAAK,EAAE,OAAO,CAAC,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;KACnD,CAAC;AACJ,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,kBAAkB,CAChC,MAAoB,EACpB,OAAmB,SAAS;IAE5B,IAAI,IAAI,KAAK,KAAK,IAAI,MAAM,CAAC,IAAI,KAAK,IAAI;QAAE,OAAO,SAAS,CAAC;IAE7D,MAAM,KAAK,GAAG,MAAM,CAAC,IAAI,KAAK,SAAS,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC;IACtD,OAAO,IAAI,KAAK,OAAO,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,GAAG,KAAK,IAAI,MAAM,CAAC,IAAI,EAAE,CAAC;AAC9D,CAAC"}
|
package/package.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
1
|
+
{
|
|
2
|
+
"name": "@mohan-cao/jev-verify",
|
|
3
|
+
"version": "0.1.0",
|
|
4
|
+
"description": "Jev verification signal: did a completed response address the request, evade it, or miss its obligation.",
|
|
5
|
+
"type": "module",
|
|
6
|
+
"license": "Apache-2.0",
|
|
7
|
+
"repository": {
|
|
8
|
+
"type": "git",
|
|
9
|
+
"url": "git+https://github.com/mohan-cao/pi-extensions.git",
|
|
10
|
+
"directory": "packages/jev-verify"
|
|
11
|
+
},
|
|
12
|
+
"keywords": [
|
|
13
|
+
"jev",
|
|
14
|
+
"typesafe-ai",
|
|
15
|
+
"system-one",
|
|
16
|
+
"classifier",
|
|
17
|
+
"llm"
|
|
18
|
+
],
|
|
19
|
+
"engines": {
|
|
20
|
+
"node": ">=20"
|
|
21
|
+
},
|
|
22
|
+
"files": [
|
|
23
|
+
"dist",
|
|
24
|
+
"README.md",
|
|
25
|
+
"LICENSE",
|
|
26
|
+
"CHANGELOG.md"
|
|
27
|
+
],
|
|
28
|
+
"exports": {
|
|
29
|
+
".": {
|
|
30
|
+
"types": "./dist/index.d.ts",
|
|
31
|
+
"default": "./dist/index.js"
|
|
32
|
+
}
|
|
33
|
+
},
|
|
34
|
+
"publishConfig": {
|
|
35
|
+
"access": "public"
|
|
36
|
+
},
|
|
37
|
+
"dependencies": {
|
|
38
|
+
"@mohan-cao/jev-classifier": "^0.2.0"
|
|
39
|
+
},
|
|
40
|
+
"scripts": {
|
|
41
|
+
"clean": "rm -rf dist",
|
|
42
|
+
"build": "tsc -p tsconfig.json",
|
|
43
|
+
"build:unchecked": "tsc -p tsconfig.json --noCheck",
|
|
44
|
+
"test": "node --test test/*.test.mjs"
|
|
45
|
+
}
|
|
46
|
+
}
|