@basein/runner 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 +276 -0
- package/dist/auth/client.d.ts +85 -0
- package/dist/auth/client.js +284 -0
- package/dist/bin/bir-hooks.d.ts +48 -0
- package/dist/bin/bir-hooks.js +201 -0
- package/dist/bin/bir-proxy.d.ts +45 -0
- package/dist/bin/bir-proxy.js +207 -0
- package/dist/bin/bir-scenario.d.ts +24 -0
- package/dist/bin/bir-scenario.js +177 -0
- package/dist/bin/bir.d.ts +21 -0
- package/dist/bin/bir.js +876 -0
- package/dist/config/adapters/claude-code.d.ts +76 -0
- package/dist/config/adapters/claude-code.js +181 -0
- package/dist/config/adapters/generic.d.ts +17 -0
- package/dist/config/adapters/generic.js +36 -0
- package/dist/config/generate.d.ts +127 -0
- package/dist/config/generate.js +114 -0
- package/dist/config/resolve.d.ts +68 -0
- package/dist/config/resolve.js +132 -0
- package/dist/control/client.d.ts +56 -0
- package/dist/control/client.js +86 -0
- package/dist/control/correlation.d.ts +86 -0
- package/dist/control/correlation.js +0 -0
- package/dist/control/discovery.d.ts +50 -0
- package/dist/control/discovery.js +123 -0
- package/dist/control/ordering.d.ts +38 -0
- package/dist/control/ordering.js +44 -0
- package/dist/control/paths.d.ts +32 -0
- package/dist/control/paths.js +56 -0
- package/dist/control/server.d.ts +272 -0
- package/dist/control/server.js +1131 -0
- package/dist/control/transcript.d.ts +75 -0
- package/dist/control/transcript.js +241 -0
- package/dist/index.d.ts +37 -0
- package/dist/index.js +32 -0
- package/dist/jsonrpc/framing.d.ts +49 -0
- package/dist/jsonrpc/framing.js +143 -0
- package/dist/jsonrpc/types.d.ts +52 -0
- package/dist/jsonrpc/types.js +46 -0
- package/dist/proxy/intercept.d.ts +55 -0
- package/dist/proxy/intercept.js +147 -0
- package/dist/proxy/relay.d.ts +97 -0
- package/dist/proxy/relay.js +166 -0
- package/dist/proxy/session.d.ts +116 -0
- package/dist/proxy/session.js +319 -0
- package/dist/record/housekeeping.d.ts +34 -0
- package/dist/record/housekeeping.js +39 -0
- package/dist/record/queue.d.ts +48 -0
- package/dist/record/queue.js +96 -0
- package/dist/record/recorder.d.ts +111 -0
- package/dist/record/recorder.js +39 -0
- package/dist/record/redact.d.ts +37 -0
- package/dist/record/redact.js +119 -0
- package/dist/record/remote-recorder.d.ts +110 -0
- package/dist/record/remote-recorder.js +301 -0
- package/dist/record/truncate.d.ts +36 -0
- package/dist/record/truncate.js +85 -0
- package/dist/replay/bundle.d.ts +36 -0
- package/dist/replay/bundle.js +89 -0
- package/dist/replay/controller.d.ts +300 -0
- package/dist/replay/controller.js +807 -0
- package/dist/replay/coverage.d.ts +41 -0
- package/dist/replay/coverage.js +56 -0
- package/dist/replay/derive.d.ts +58 -0
- package/dist/replay/derive.js +166 -0
- package/dist/replay/executor.d.ts +78 -0
- package/dist/replay/executor.js +233 -0
- package/dist/replay/logic.d.ts +31 -0
- package/dist/replay/logic.js +50 -0
- package/dist/replay/plan.d.ts +181 -0
- package/dist/replay/plan.js +397 -0
- package/dist/replay/pricing.d.ts +41 -0
- package/dist/replay/pricing.js +76 -0
- package/dist/replay/source-run.d.ts +50 -0
- package/dist/replay/source-run.js +98 -0
- package/dist/replay/tool-error.d.ts +22 -0
- package/dist/replay/tool-error.js +60 -0
- package/dist/replay/types.d.ts +116 -0
- package/dist/replay/types.js +35 -0
- package/dist/upstream/client.d.ts +78 -0
- package/dist/upstream/client.js +114 -0
- package/dist/upstream/http-client.d.ts +78 -0
- package/dist/upstream/http-client.js +261 -0
- package/dist/upstream/lazy-client.d.ts +31 -0
- package/dist/upstream/lazy-client.js +53 -0
- package/dist/upstream/stdio-client.d.ts +57 -0
- package/dist/upstream/stdio-client.js +203 -0
- package/dist/util/log.d.ts +27 -0
- package/dist/util/log.js +51 -0
- package/dist/util/version.d.ts +2 -0
- package/dist/util/version.js +40 -0
- package/docs/BaseInstRunner.md +621 -0
- package/docs/calculatedReplay.md +1185 -0
- package/docs/calculatedReplayGuide.md +448 -0
- package/docs/installRun.md +413 -0
- package/docs/mcpmark.md +752 -0
- package/docs/quickstart.md +201 -0
- package/docs/t-bench.md +394 -0
- package/package.json +56 -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 +1,276 @@
|
|
|
1
|
+
# BaseInstRunnerMCP
|
|
2
|
+
|
|
3
|
+
A recording MCP **proxy**. It sits between any MCP client (Claude Code, Cursor,
|
|
4
|
+
Codex, an Agent SDK session) and every MCP server that client uses, executes each
|
|
5
|
+
call on the client's behalf, and records the run as a reusable scenario in the
|
|
6
|
+
BaseIn service.
|
|
7
|
+
|
|
8
|
+
The design this implements is [docs/BaseInstRunner.md](docs/BaseInstRunner.md).
|
|
9
|
+
This README is the operator's half: how to run it, what it records, and what it
|
|
10
|
+
cannot. For v2 — running a calculated scenario when a prompt means the same thing
|
|
11
|
+
as one you have run before — see [docs/calculatedReplay.md](docs/calculatedReplay.md)
|
|
12
|
+
and its runbook [docs/calculatedReplayGuide.md](docs/calculatedReplayGuide.md).
|
|
13
|
+
|
|
14
|
+
```
|
|
15
|
+
model ──mcp__chrome-devtools__*──▶ bir-proxy ──stdio/http──▶ chrome-devtools-mcp
|
|
16
|
+
hooks ──built-ins, prompt, stop──▶ bir-hooks (control server) ──▶ BaseIn /recordings/*
|
|
17
|
+
▲
|
|
18
|
+
bir-proxy ────┘ step reports
|
|
19
|
+
```
|
|
20
|
+
|
|
21
|
+
---
|
|
22
|
+
|
|
23
|
+
## Install
|
|
24
|
+
|
|
25
|
+
On a runner machine, one command does machine, account and project scope:
|
|
26
|
+
|
|
27
|
+
```powershell
|
|
28
|
+
.\scripts\install-runner.ps1 C:\work\repo -AuthUrl https://your-basein-service # Windows
|
|
29
|
+
```
|
|
30
|
+
|
|
31
|
+
```bash
|
|
32
|
+
BIR_AUTH_URL=https://your-basein-service ./scripts/install-runner.sh ~/work/repo # macOS/Linux
|
|
33
|
+
```
|
|
34
|
+
|
|
35
|
+
Step by step, for one machine: [docs/quickstart.md](docs/quickstart.md).
|
|
36
|
+
For a fleet — packaging, releases, offline installs, updates:
|
|
37
|
+
[docs/installRun.md](docs/installRun.md).
|
|
38
|
+
|
|
39
|
+
By hand, or when developing BaseInstRunner itself:
|
|
40
|
+
|
|
41
|
+
```bash
|
|
42
|
+
npm i -g @basein/runner # or: npm install && npm run build, in a checkout
|
|
43
|
+
cd the-project-you-record-from
|
|
44
|
+
bir install --global # --local from a checkout; plain `bir install` writes a pinned npx entry
|
|
45
|
+
```
|
|
46
|
+
|
|
47
|
+
`bir install` rewrites your MCP entries in place, keeping every key (so
|
|
48
|
+
`mcp__<server>__<tool>` names never change), and wires Claude Code's hooks. It
|
|
49
|
+
stashes the original text of every file first — `bir uninstall` puts it back
|
|
50
|
+
byte-for-byte.
|
|
51
|
+
|
|
52
|
+
```bash
|
|
53
|
+
export BIR_AUTH_URL=https://your-basein-service # required to record anything
|
|
54
|
+
bir login # once
|
|
55
|
+
bir-hooks # leave running; Ctrl-C to stop
|
|
56
|
+
claude # your session, now recorded
|
|
57
|
+
```
|
|
58
|
+
|
|
59
|
+
Check it:
|
|
60
|
+
|
|
61
|
+
```bash
|
|
62
|
+
bir status # what is wrapped, and which config scope won
|
|
63
|
+
bir doctor # is it working *right now* — exits non-zero if not
|
|
64
|
+
```
|
|
65
|
+
|
|
66
|
+
### Any other MCP client
|
|
67
|
+
|
|
68
|
+
Nothing here knows where Cursor or Codex keep their settings, and it does not
|
|
69
|
+
guess. Two paths:
|
|
70
|
+
|
|
71
|
+
```bash
|
|
72
|
+
bir install --config /path/to/that/client/mcp.json
|
|
73
|
+
bir wrap --server-name my-server -- node my-server.js # prints an entry to paste
|
|
74
|
+
```
|
|
75
|
+
|
|
76
|
+
---
|
|
77
|
+
|
|
78
|
+
## The two tiers
|
|
79
|
+
|
|
80
|
+
Only Claude Code has hooks, so run ownership is negotiated at proxy startup.
|
|
81
|
+
|
|
82
|
+
| Tier | When | Records |
|
|
83
|
+
|---|---|---|
|
|
84
|
+
| **1 — Bound** | a control server (`bir-hooks`) is discoverable for this cwd | built-ins **and** MCP, one ordered step stream, prompt and final answer |
|
|
85
|
+
| **2 — Standalone** | no control server within 5 s | MCP calls only — no prompt, no final answer, no built-in steps |
|
|
86
|
+
|
|
87
|
+
Tier 2 is not a bug; it is the honest ceiling of what an MCP proxy can observe. A
|
|
88
|
+
run records which tier produced it in `metadata.tier`, so nothing downstream
|
|
89
|
+
mistakes a partial trace for a complete one.
|
|
90
|
+
|
|
91
|
+
Steps taken while the tier is still being decided are **buffered**, not dropped —
|
|
92
|
+
the host may spawn MCP servers before `SessionStart` fires, and the ordering
|
|
93
|
+
between the two is not guaranteed.
|
|
94
|
+
|
|
95
|
+
---
|
|
96
|
+
|
|
97
|
+
## What it will not do
|
|
98
|
+
|
|
99
|
+
- **Non-MCP tools without the hook binary.** `Bash`, `Read`, `Edit` and friends
|
|
100
|
+
are not MCP traffic. An MCP server sees only MCP traffic addressed to it. That
|
|
101
|
+
is a structural fact, not a limitation of this implementation.
|
|
102
|
+
- **`claude-in-chrome`.** It is `scope: "dynamic"`, present in no config file, so
|
|
103
|
+
there is no entry to rewrite.
|
|
104
|
+
- **Replay, unless you turn it on.** Recording is the default and the whole of
|
|
105
|
+
what runs out of the box. Calculated replay — running a matched prompt's
|
|
106
|
+
scenario instead of rediscovering it — ships, but needs `bir install --replay`
|
|
107
|
+
*and* `BIR_REPLAY=1`, for the reason in the next section.
|
|
108
|
+
- **Change what the model does.** Tool descriptions are never edited. The
|
|
109
|
+
model-visible changes are schema relaxation, below, and — only with
|
|
110
|
+
`--replay` — one first-party tool.
|
|
111
|
+
|
|
112
|
+
---
|
|
113
|
+
|
|
114
|
+
## Calculated replay
|
|
115
|
+
|
|
116
|
+
When the service recognises that a prompt means the same thing as one you have
|
|
117
|
+
run before, it hands back the *calculated scenario* derived from that earlier
|
|
118
|
+
run: an intent, a parameter schema, and one bit of logic per step that recomputes
|
|
119
|
+
that step's arguments. `bir` can run it instead of letting the model rediscover
|
|
120
|
+
the plan token by token.
|
|
121
|
+
|
|
122
|
+
The design is [docs/calculatedReplay.md](docs/calculatedReplay.md); the runbook —
|
|
123
|
+
record, calculate, check, enable, read the audit lines — is
|
|
124
|
+
[docs/calculatedReplayGuide.md](docs/calculatedReplayGuide.md).
|
|
125
|
+
|
|
126
|
+
```bash
|
|
127
|
+
bir install --replay # adds the `bir` MCP server; raises the prompt-hook timeout
|
|
128
|
+
BIR_REPLAY=1 bir-hooks # and this is the switch
|
|
129
|
+
```
|
|
130
|
+
|
|
131
|
+
Two switches, not one, because of what replay is: **a scenario is a pre-approved
|
|
132
|
+
list of tool calls with computed arguments.** A steered step is answered
|
|
133
|
+
`permissionDecision: "allow"`, so a pinned `Bash` or `Edit` runs without the
|
|
134
|
+
prompt you would have seen. A *directly executed* step never reaches the
|
|
135
|
+
permission system at all — it is dispatched to a `bir-proxy` and called on a
|
|
136
|
+
connection that is already open. Set `BIR_REPLAY_ALLOW_SERVERS` to the servers
|
|
137
|
+
you are content to have called unattended, and read
|
|
138
|
+
[the guide's §5.1](docs/calculatedReplayGuide.md) before enabling it anywhere
|
|
139
|
+
that matters.
|
|
140
|
+
|
|
141
|
+
What it buys: on a scenario whose every step is a wrapped MCP tool, the whole
|
|
142
|
+
sequence runs through connections the proxies already hold — **zero model tokens**
|
|
143
|
+
— and the model's only job is one tool call that reads the results. Mixed
|
|
144
|
+
scenarios steer the live session instead, pinning each step's arguments. Either
|
|
145
|
+
way the turn reports what it cost, so the savings ledger measures something real.
|
|
146
|
+
|
|
147
|
+
---
|
|
148
|
+
|
|
149
|
+
## Two things to know before you turn it on
|
|
150
|
+
|
|
151
|
+
**1. Correlation auto-approves wrapped MCP calls.** Merging the hook's view of a
|
|
152
|
+
call with the proxy's requires carrying an id through the tool's arguments, and
|
|
153
|
+
the mechanism for that is answering `PreToolUse` with
|
|
154
|
+
`permissionDecision: "allow"`. That skips a permission prompt you might otherwise
|
|
155
|
+
have seen for those tools. To keep the prompt:
|
|
156
|
+
|
|
157
|
+
```bash
|
|
158
|
+
BIR_CORRELATION_DECISION=ask bir-hooks
|
|
159
|
+
```
|
|
160
|
+
|
|
161
|
+
Hosts that ignore `updatedInput` on a non-allow decision then fall back to
|
|
162
|
+
fingerprint matching, which is lossy under identical concurrent calls.
|
|
163
|
+
|
|
164
|
+
**2. Schema relaxation is visible to the model.** Carrying that id means adding
|
|
165
|
+
`additionalProperties: true` plus one declared key to each wrapped tool's schema,
|
|
166
|
+
on every call — not only recorded ones. It is applied **only** in Tier 1 with
|
|
167
|
+
correlation active, and **only** to wrapped servers. `--no-correlation` turns it
|
|
168
|
+
off entirely and joins on `(server, tool, hash(args))` inside a 30 s window
|
|
169
|
+
instead.
|
|
170
|
+
|
|
171
|
+
---
|
|
172
|
+
|
|
173
|
+
## Configuration
|
|
174
|
+
|
|
175
|
+
| Variable | Effect |
|
|
176
|
+
|---|---|
|
|
177
|
+
| `BIR_AUTH_URL` | BaseIn auth-service. **Unset means nothing is recorded** — everything else still works. |
|
|
178
|
+
| `BIR_AUTH_DISABLE=1` | Skip authentication (local dev / CI). |
|
|
179
|
+
| `BIR_CONTROL_PORT` | Preferred control-server port (default 53411). |
|
|
180
|
+
| `BIR_CONTROL_URL` / `BIR_CONTROL_TOKEN` | Point a proxy at a control server directly, bypassing discovery. Injected into generated config; the only reliable channel for an SDK session. |
|
|
181
|
+
| `BIR_CORRELATION_DECISION` | `allow` (default) or `ask` — see above. |
|
|
182
|
+
| `BIR_NO_CORRELATION=1` | Never inject a call id. |
|
|
183
|
+
| `BIR_HOME` | Where state lives (default `~/.baseinstrunner`). |
|
|
184
|
+
| `BIR_CLAUDE_JSON` | Override the `~/.claude.json` location. |
|
|
185
|
+
| `BIR_VERBOSE=1` | Per-message detail lines on stderr. |
|
|
186
|
+
|
|
187
|
+
Calculated replay adds these, all read by `bir-hooks` only — the proxies are told
|
|
188
|
+
what to do when they register, so there is one switch and no second copy of it:
|
|
189
|
+
|
|
190
|
+
| Variable | Effect |
|
|
191
|
+
|---|---|
|
|
192
|
+
| `BIR_REPLAY=1` | Enable replay. Nothing below matters until it is set. |
|
|
193
|
+
| `BIR_REPLAY_ALLOW_SERVERS` | Server keys eligible for **direct** execution. Unset means every wrapped server; setting it is the recommendation. |
|
|
194
|
+
| `BIR_MIN_STEER_SIMILARITY` | Minimum match similarity to replay (default `0.92`, above the service's own `0.9` detection threshold). |
|
|
195
|
+
| `ANTHROPIC_API_KEY` | Enables parameter derivation. Without it, replay uses the scenario's recorded sample values — free, and often still correct. |
|
|
196
|
+
| `BIR_DERIVE_MODEL` | Derivation model (default `claude-haiku-4-5-20251001`). |
|
|
197
|
+
| `BIR_MATCH_BUDGET_MS` | How long the prompt hook waits for a match (default `2500`). |
|
|
198
|
+
| `BIR_DERIVE_BUDGET_MS` | How long the first `PreToolUse` waits for parameters (default `8000`). |
|
|
199
|
+
| `BIR_REPLAY_BUDGET_MS` / `BIR_STEP_TIMEOUT_MS` | Whole-plan and per-step ceilings for direct execution (default `120000` / `60000`). |
|
|
200
|
+
|
|
201
|
+
### The audit log
|
|
202
|
+
|
|
203
|
+
Every decision is one `key=value` line on **stderr** — never stdout, which is the
|
|
204
|
+
host's JSON-RPC stream. There is no file sink to configure; persistence is the
|
|
205
|
+
shell's job:
|
|
206
|
+
|
|
207
|
+
```bash
|
|
208
|
+
bir-hooks 2>&1 | tee -a ~/.baseinstrunner/audit.log
|
|
209
|
+
grep run.lossy ~/.baseinstrunner/audit.log
|
|
210
|
+
```
|
|
211
|
+
|
|
212
|
+
---
|
|
213
|
+
|
|
214
|
+
## Failure behaviour
|
|
215
|
+
|
|
216
|
+
The governing rule: **a host session never fails because of BaseInstRunner.**
|
|
217
|
+
|
|
218
|
+
| Failure | Behaviour |
|
|
219
|
+
|---|---|
|
|
220
|
+
| Control server unreachable at startup | buffer ≤ 5 s, then Tier 2; logged once |
|
|
221
|
+
| Control server dies mid-session | the proxy switches to Tier 2 for the rest; run flagged lossy |
|
|
222
|
+
| Recorder HTTP fails | send dropped and logged; never retried inline |
|
|
223
|
+
| Upstream fails to start | `initialize` still answers; requests return an error — one broken server, not a broken session |
|
|
224
|
+
| Upstream dies mid-call | `isError: true` with the transport message, so the model can react |
|
|
225
|
+
| Unknown JSON-RPC method | forwarded; any rejection is relayed back |
|
|
226
|
+
| Step queue overflow | oldest dropped, run flagged lossy |
|
|
227
|
+
| Host sends unparseable input | `-32700` and the relay keeps going |
|
|
228
|
+
|
|
229
|
+
---
|
|
230
|
+
|
|
231
|
+
## Development
|
|
232
|
+
|
|
233
|
+
```bash
|
|
234
|
+
npm run build # src → dist
|
|
235
|
+
npm test # 233 tests: transcript identity, passthrough, correlation, config, replay
|
|
236
|
+
npm run typecheck
|
|
237
|
+
npm run test:smoke # end-to-end replay with real processes (needs build + pretest first)
|
|
238
|
+
```
|
|
239
|
+
|
|
240
|
+
The core test is **transcript identity**: the same scripted conversation is run
|
|
241
|
+
against a fake upstream directly and through the proxy, and every host-visible
|
|
242
|
+
message must match. If a change breaks transparency, that test says so first.
|
|
243
|
+
|
|
244
|
+
`test/smoke-replay.mjs` is the other one worth running by hand. Everything else
|
|
245
|
+
runs in-process; this starts a stand-in BaseIn service, a real `bir-hooks`, a real
|
|
246
|
+
`bir-proxy` wrapping a real MCP server and a real `bir-scenario`, then drives a
|
|
247
|
+
two-step scenario through actual stdio. It is what catches the seams between
|
|
248
|
+
processes — and it is how the "a proxy between polls is not a missing proxy" bug
|
|
249
|
+
was found.
|
|
250
|
+
|
|
251
|
+
`test/fake-upstream.ts` is a scriptable MCP server — arbitrary capabilities,
|
|
252
|
+
prompts, resources, `list_changed` notifications, reverse-direction
|
|
253
|
+
`elicitation/create`, delays, failures, and dying on command. Everything else
|
|
254
|
+
builds on it.
|
|
255
|
+
|
|
256
|
+
---
|
|
257
|
+
|
|
258
|
+
## Known gaps
|
|
259
|
+
|
|
260
|
+
- **`metadata.lossy` is the value known at run start.** A run that turns lossy
|
|
261
|
+
later says so on the audit log (`run.lossy`) and in the step gap, but the
|
|
262
|
+
metadata is not updated: the service exposes no metadata patch, and this is
|
|
263
|
+
exactly open question 1 in the design. Fixing it properly needs a server-side
|
|
264
|
+
decision, not a client-side workaround.
|
|
265
|
+
- **Subagent steps are attributed, not partitioned.** `SubagentStart` /
|
|
266
|
+
`SubagentStop` give us `agent_id`, which is prefixed onto a step's `context`.
|
|
267
|
+
A real agent column is a schema change, and out of scope for v1 (D6).
|
|
268
|
+
- **`--lazy` defers to the first host message, not to `tools/list`.** Deferring
|
|
269
|
+
further would mean answering `initialize` before the upstream exists — that is,
|
|
270
|
+
fabricating its capabilities, which is the thing Phase 3 exists to prevent. For
|
|
271
|
+
a client that connects at session start (Claude Code does) the flag saves
|
|
272
|
+
nothing; for one that connects on demand it saves the whole process. See
|
|
273
|
+
`src/upstream/lazy-client.ts`.
|
|
274
|
+
- **Coexistence with RRepeat is untested.** Both would record the same session.
|
|
275
|
+
Runs carry `metadata.recorder`, so server-side dedupe is possible, but nothing
|
|
276
|
+
here detects the other (design §13, open question 2).
|
|
@@ -0,0 +1,85 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* auth client — ported from RRepeat's `src/auth/client.ts` (Phase 4.1).
|
|
3
|
+
*
|
|
4
|
+
* Authenticates the CLI user against the BaseIn auth-service and returns a JWT
|
|
5
|
+
* session. Every recording call carries that access token; the recording's owner
|
|
6
|
+
* is the JWT subject on the server side, which is what makes tenancy work
|
|
7
|
+
* without BaseInstRunner knowing anything about it.
|
|
8
|
+
*
|
|
9
|
+
* Flow:
|
|
10
|
+
* 1. Load the cached session from ~/.baseinstrunner/credentials.json.
|
|
11
|
+
* 2. If the access token is still valid, reuse it.
|
|
12
|
+
* 3. Else if a refresh token is present, POST /auth/refresh (silent).
|
|
13
|
+
* 4. Else prompt for email + password and POST /auth/login.
|
|
14
|
+
* 5. Persist the result (mode 0600) and return it.
|
|
15
|
+
*
|
|
16
|
+
* WHY IT NEVER `process.exit`s: `bir-proxy` runs inside the host's process tree.
|
|
17
|
+
* Failing to authenticate must degrade to not-recording (§10), not kill a server
|
|
18
|
+
* the host is waiting on. Callers get `undefined` and decide.
|
|
19
|
+
*
|
|
20
|
+
* Config:
|
|
21
|
+
* BIR_AUTH_URL base URL of the BaseIn auth-service (required).
|
|
22
|
+
* BIR_AUTH_DISABLE "1" to skip auth entirely (local dev / CI).
|
|
23
|
+
*/
|
|
24
|
+
export interface AuthUser {
|
|
25
|
+
id: string;
|
|
26
|
+
email: string;
|
|
27
|
+
name?: string;
|
|
28
|
+
}
|
|
29
|
+
export interface AuthSession {
|
|
30
|
+
accessToken: string;
|
|
31
|
+
refreshToken: string;
|
|
32
|
+
/** Epoch milliseconds at which the access token expires. */
|
|
33
|
+
accessExpiresAt: number;
|
|
34
|
+
user: AuthUser;
|
|
35
|
+
}
|
|
36
|
+
export interface AuthenticateOptions {
|
|
37
|
+
/** Base URL of the auth service. Defaults to BIR_AUTH_URL. */
|
|
38
|
+
authUrl?: string;
|
|
39
|
+
/** Never prompt: fail (return undefined) if no valid/refreshable session exists. */
|
|
40
|
+
nonInteractive?: boolean;
|
|
41
|
+
}
|
|
42
|
+
/**
|
|
43
|
+
* Where the cached session lives.
|
|
44
|
+
*
|
|
45
|
+
* A **function**, not a constant: `configDir()` reads `BIR_HOME`, and freezing
|
|
46
|
+
* this at import time meant the directory we created and the file we wrote could
|
|
47
|
+
* disagree the moment anything set that variable after load — which surfaced as
|
|
48
|
+
* an ENOENT deep inside a token refresh.
|
|
49
|
+
*/
|
|
50
|
+
export declare function credentialsPath(): string;
|
|
51
|
+
/** Load the cached session, or undefined if missing/unreadable/malformed. */
|
|
52
|
+
export declare function loadCredentials(): AuthSession | undefined;
|
|
53
|
+
/** Persist the session with mode 0600. */
|
|
54
|
+
export declare function saveCredentials(session: AuthSession): void;
|
|
55
|
+
/** Remove the cached session (`bir logout`). No-op if absent. */
|
|
56
|
+
export declare function clearCredentials(): void;
|
|
57
|
+
export declare function resolveAuthUrl(opts?: AuthenticateOptions): string;
|
|
58
|
+
/**
|
|
59
|
+
* Follow a redirect on the base URL *once*, so a bearer token is never sent to
|
|
60
|
+
* a URL that will bounce it.
|
|
61
|
+
*
|
|
62
|
+
* WHY THIS EXISTS. A service behind a TLS-terminating proxy answers plain HTTP
|
|
63
|
+
* with `308 → https://…`. Node's `fetch` follows that, and because `http:` and
|
|
64
|
+
* `https:` are different origins the spec requires it to **strip the
|
|
65
|
+
* `Authorization` header**. The request arrives unauthenticated.
|
|
66
|
+
*
|
|
67
|
+
* The failure is maddening because sign-in works: login sends credentials in the
|
|
68
|
+
* BODY, which a 308 preserves. So `auth.ok` is printed, and then every
|
|
69
|
+
* subsequent call fails `401 {"error":"missing_token"}` — the server saying it
|
|
70
|
+
* saw no token at all, which reads like a bug in the client rather than a
|
|
71
|
+
* redirect. Observed against a Caddy-fronted deployment; the whole session
|
|
72
|
+
* recorded nothing.
|
|
73
|
+
*
|
|
74
|
+
* A same-host upgrade to https is adopted silently — it is what the operator
|
|
75
|
+
* meant. Anything else (a different host, a downgrade to http) is reported and
|
|
76
|
+
* NOT followed: quietly re-sending someone's bearer token to a host they did not
|
|
77
|
+
* configure is worse than failing.
|
|
78
|
+
*/
|
|
79
|
+
export declare function normalizeAuthUrl(baseUrl: string, fetchImpl?: typeof fetch): Promise<string>;
|
|
80
|
+
/**
|
|
81
|
+
* Authenticate and return a valid session, or `undefined` when none can be
|
|
82
|
+
* obtained. Callers degrade to not-recording rather than failing the session.
|
|
83
|
+
*/
|
|
84
|
+
export declare function authenticate(opts?: AuthenticateOptions): Promise<AuthSession | undefined>;
|
|
85
|
+
//# sourceMappingURL=client.d.ts.map
|