dvalincode 0.17.0 โ†’ 0.19.0

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.
Files changed (104) hide show
  1. package/README.md +320 -46
  2. package/README.zh-CN.md +277 -39
  3. package/THIRD_PARTY_NOTICES.md +98 -43
  4. package/dist/cli.js +4 -0
  5. package/dist/cli.js.map +1 -1
  6. package/dist/commands/dvalin.d.ts +41 -1
  7. package/dist/commands/dvalin.js +219 -37
  8. package/dist/commands/dvalin.js.map +1 -1
  9. package/dist/commands/mcpInstall.d.ts +2 -0
  10. package/dist/commands/mcpInstall.js +69 -0
  11. package/dist/commands/mcpInstall.js.map +1 -0
  12. package/dist/commands/run.js +21 -0
  13. package/dist/commands/run.js.map +1 -1
  14. package/dist/commands/security.d.ts +47 -0
  15. package/dist/commands/security.js +415 -0
  16. package/dist/commands/security.js.map +1 -0
  17. package/dist/dvalinCli.d.ts +3 -0
  18. package/dist/dvalinCli.js +36 -0
  19. package/dist/dvalinCli.js.map +1 -0
  20. package/dist/dvalinIndex.d.ts +2 -0
  21. package/dist/dvalinIndex.js +9 -0
  22. package/dist/dvalinIndex.js.map +1 -0
  23. package/dist/evidence/pack.d.ts +8 -0
  24. package/dist/evidence/pack.js +16 -0
  25. package/dist/evidence/pack.js.map +1 -1
  26. package/dist/harness/run.d.ts +15 -0
  27. package/dist/harness/run.js +45 -19
  28. package/dist/harness/run.js.map +1 -1
  29. package/dist/harness/verification.d.ts +56 -0
  30. package/dist/harness/verification.js +90 -0
  31. package/dist/harness/verification.js.map +1 -0
  32. package/dist/mcp/install.d.ts +48 -0
  33. package/dist/mcp/install.js +74 -0
  34. package/dist/mcp/install.js.map +1 -0
  35. package/dist/mcp/server.d.ts +10 -1
  36. package/dist/mcp/server.js +251 -34
  37. package/dist/mcp/server.js.map +1 -1
  38. package/dist/remediation/automate.d.ts +4 -1
  39. package/dist/remediation/automate.js +35 -18
  40. package/dist/remediation/automate.js.map +1 -1
  41. package/dist/remediation/diffScope.d.ts +29 -0
  42. package/dist/remediation/diffScope.js +107 -0
  43. package/dist/remediation/diffScope.js.map +1 -0
  44. package/dist/remediation/executor.d.ts +60 -0
  45. package/dist/remediation/executor.js +144 -0
  46. package/dist/remediation/executor.js.map +1 -0
  47. package/dist/remediation/localScan.d.ts +2 -1
  48. package/dist/remediation/localScan.js +73 -3
  49. package/dist/remediation/localScan.js.map +1 -1
  50. package/dist/remediation/sarif.d.ts +6 -0
  51. package/dist/remediation/sarif.js +36 -14
  52. package/dist/remediation/sarif.js.map +1 -1
  53. package/dist/remediation/scannerSuite.d.ts +21 -0
  54. package/dist/remediation/scannerSuite.js +65 -4
  55. package/dist/remediation/scannerSuite.js.map +1 -1
  56. package/dist/remediation/verify.d.ts +38 -0
  57. package/dist/remediation/verify.js +116 -0
  58. package/dist/remediation/verify.js.map +1 -0
  59. package/dist/security/baseline.d.ts +7 -0
  60. package/dist/security/baseline.js +26 -0
  61. package/dist/security/baseline.js.map +1 -0
  62. package/dist/security/config.d.ts +32 -0
  63. package/dist/security/config.js +131 -0
  64. package/dist/security/config.js.map +1 -0
  65. package/dist/security/contracts.d.ts +149 -0
  66. package/dist/security/contracts.js +231 -0
  67. package/dist/security/contracts.js.map +1 -0
  68. package/dist/security/fixRecord.d.ts +231 -0
  69. package/dist/security/fixRecord.js +327 -0
  70. package/dist/security/fixRecord.js.map +1 -0
  71. package/dist/security/fixRecordFile.d.ts +8 -0
  72. package/dist/security/fixRecordFile.js +37 -0
  73. package/dist/security/fixRecordFile.js.map +1 -0
  74. package/dist/security/fixRecordStore.d.ts +23 -0
  75. package/dist/security/fixRecordStore.js +74 -0
  76. package/dist/security/fixRecordStore.js.map +1 -0
  77. package/dist/security/render.d.ts +8 -0
  78. package/dist/security/render.js +20 -0
  79. package/dist/security/render.js.map +1 -0
  80. package/dist/security/suppressions.d.ts +12 -0
  81. package/dist/security/suppressions.js +60 -0
  82. package/dist/security/suppressions.js.map +1 -0
  83. package/dist/security/verifyRun.d.ts +21 -0
  84. package/dist/security/verifyRun.js +74 -0
  85. package/dist/security/verifyRun.js.map +1 -0
  86. package/dist/security/workflow.d.ts +98 -0
  87. package/dist/security/workflow.js +276 -0
  88. package/dist/security/workflow.js.map +1 -0
  89. package/dist/server/routes/remediation.d.ts +19 -0
  90. package/dist/server/routes/remediation.js +90 -18
  91. package/dist/server/routes/remediation.js.map +1 -1
  92. package/dist/tools/runCheck.d.ts +6 -0
  93. package/dist/tools/runCheck.js +1 -1
  94. package/dist/tools/runCheck.js.map +1 -1
  95. package/dist/tools/runSecuritySuite.d.ts +3 -0
  96. package/dist/tools/runSecuritySuite.js +18 -11
  97. package/dist/tools/runSecuritySuite.js.map +1 -1
  98. package/dist/tui/app.js +21 -2
  99. package/dist/tui/app.js.map +1 -1
  100. package/dist/tui/render.js +2 -0
  101. package/dist/tui/render.js.map +1 -1
  102. package/dist/version.d.ts +1 -1
  103. package/dist/version.js +1 -1
  104. package/package.json +17 -10
package/README.md CHANGED
@@ -9,7 +9,7 @@
9
9
  <p align="center">
10
10
  <a href="https://github.com/arthurpanhku/dvalincode/releases/latest"><img src="https://img.shields.io/github/v/release/arthurpanhku/dvalincode?style=for-the-badge&color=818cf8&label=Release" alt="Release"></a>
11
11
  <a href="https://github.com/arthurpanhku/dvalincode/releases"><img src="https://img.shields.io/github/downloads/arthurpanhku/dvalincode/total?style=for-the-badge&color=blue&label=Downloads" alt="Downloads"></a>
12
- <a href="#-tests"><img src="https://img.shields.io/badge/Tests-357%20%2F%20357%20%E2%9C%93-success?style=for-the-badge" alt="Tests"></a>
12
+ <a href="#-tests"><img src="https://img.shields.io/badge/Tests-584%20%2F%20584%20%E2%9C%93-success?style=for-the-badge" alt="Tests"></a>
13
13
  <a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-green?style=for-the-badge" alt="License"></a>
14
14
  <a href="https://scorecard.dev/viewer/?uri=github.com/arthurpanhku/dvalincode"><img src="https://api.scorecard.dev/projects/github.com/arthurpanhku/dvalincode/badge" alt="OpenSSF Scorecard"></a>
15
15
  <a href="#-quick-install"><img src="https://img.shields.io/badge/Platforms-macOS%20ยท%20Windows%20ยท%20Linux-blue?style=for-the-badge" alt="Platforms"></a>
@@ -18,23 +18,90 @@
18
18
  </p>
19
19
 
20
20
  <p align="center">
21
- <b>Find the security holes in your repo, fix them, and prove the fix โ€” in one command.</b><br>
22
- Every fix is diffed, tested, re-scanned, and recorded in a tamper-evident audit log before it can become a PR.
21
+ <b>Open security engineering for code written by humans and AI agents.</b><br>
22
+ <b>Every repair carries its own proof.</b>
23
23
  </p>
24
24
 
25
+ When an agent fixes a security finding, someone has to decide whether the fix
26
+ worked. Almost every tool asks the model that wrote it โ€” which is the one
27
+ question a model cannot answer against its own interest.
28
+
29
+ **Dvalin decides instead, and hands you the proof.** It re-scans, runs your
30
+ project's own tests itself, and reads the exit codes from processes it started.
31
+ Who wrote the repair โ€” our agent, Claude Code, Codex, Copilot, a person โ€” is
32
+ recorded and never consulted. The result is a **Verified Fix Record**: a small
33
+ JSON file anyone can re-check offline, on a laptop with no network and no
34
+ Dvalin state.
35
+
36
+ ```sh
37
+ dvalin verify-fix fix-record.json
38
+ ```
39
+ ```
40
+ Fix record 2c9d71ac03e0 ยท VERIFIED ยท scan-and-checks
41
+ executor: claude-code (recorded, not consulted)
42
+ targets: 1 before ยท 0 remaining
43
+ coverage: complete โ†’ complete
44
+ introduced: 0 (gate high/new)
45
+ outcome: verified
46
+ โœ“ test: npm run test (exit 0)
47
+ audit: run verify-36509f42 @ 414644c75af0
48
+ ```
49
+
50
+ That record says something narrow on purpose: *these findings were gone, and
51
+ these checks were observed to pass.* It is not a claim that your code is safe,
52
+ and Dvalin will not let it be read as one โ€” every record carries what the scan
53
+ actually covered, and a repair no check could confirm does not pass.
54
+ [The open profile โ†’](docs/spec/FIX-VERIFICATION.md)
55
+
56
+ A repair is a change, and a change can add as well as remove. So the record also
57
+ carries what the re-scan saw that the first scan did not, and the gate threshold
58
+ the verdict was reached under: a fix that removes an `eval` and introduces an SQL
59
+ injection is recorded as `regressed` and does not verify. Neither does a record
60
+ whose issuer never looked โ€” `introduced: not determined` fails, because a
61
+ verifier that skips the question must not score better than one that asks it and
62
+ finds something.
63
+
64
+ Dvalin is the independent security runtime between code generation and merge.
65
+ Humans, coding agents, and CI call the same versioned contract for discovery,
66
+ remediation, and verification. It runs independently, or interoperates with
67
+ specialist systems such as Codex Security through portable SARIF. Its built-in
68
+ coding capability is a remediation executor โ€” not the trust boundary, and not an
69
+ attempt to compete with every general-purpose coding agent. See the
70
+ [security-agent strategy](docs/SECURITY-AGENT-STRATEGY.md).
71
+
25
72
  ---
26
73
 
27
74
  ## โฑ๏ธ 30 seconds, no install, no API key
28
75
 
29
76
  ```sh
30
- npx dvalincode dvalin .
77
+ npx dvalincode security scan .
78
+ # After installing the package: dvalin scan .
31
79
  ```
32
80
 
33
81
  That is the whole thing. It runs the built-in rules for injection, hardcoded
34
82
  secrets, XSS, `eval`, and unsafe shell use against the current directory and
35
83
  prints what it found. No account, no model, no config, no code leaves your
36
- machine. Add `--scanners builtin,semgrep,trivy,osv-scanner` to pull in whichever
37
- of those engines you already have on `PATH`.
84
+ machine. The default policy runs only Dvalin Built-in, so the first scan always
85
+ works. Add optional engines explicitly, or inspect their fixed install commands:
86
+
87
+ ```sh
88
+ dvalin scanners list
89
+ dvalin scanners install semgrep # review the command
90
+ dvalin scanners install semgrep --yes # execute it under Dvalin policy
91
+ ```
92
+
93
+ For an incremental โ€œno new high-risk findingsโ€ gate, commit the policy and
94
+ baseline with the repository:
95
+
96
+ ```sh
97
+ dvalin init
98
+ dvalin baseline
99
+ dvalin scan
100
+ ```
101
+
102
+ This creates `dvalin.security.json` and `.dvalin/baseline.json`. Suppressions
103
+ require a reason and may have an owner and expiry date. Scan output is a
104
+ versioned envelope with a deterministic gate result and a resumable workflow ID.
38
105
 
39
106
  ### Or put it on every pull request โ€” nothing to install at all
40
107
 
@@ -45,15 +112,63 @@ permissions:
45
112
  security-events: write
46
113
  steps:
47
114
  - uses: actions/checkout@v5
48
- - uses: arthurpanhku/dvalincode@v0.17.0
115
+ with:
116
+ fetch-depth: 0 # so the scan can reach the base commit
117
+ - uses: arthurpanhku/dvalincode@v0.18.0
49
118
  with:
50
119
  fail-on: high
120
+ diff: true # only report on what this PR changed
51
121
  ```
52
122
 
53
123
  Findings land inline on the pull request diff and in your Security tab.
54
124
  No API key, no secrets, no model โ€” the scan is deterministic and local to the
55
125
  runner. [Full example โ†’](docs/examples/dvalin-scan.yml)
56
126
 
127
+ `diff: true` reports only on lines the pull request changed, so the gate blocks
128
+ what this change *adds* instead of everything the repository already carried.
129
+ That is what makes the check adoptable on a codebase that was not clean to
130
+ begin with. Drop it to scan the whole repository.
131
+
132
+ Every comment states what the scan **covered** โ€” `complete`, `partial`, or
133
+ `unknown` โ€” beside the result, because "no findings" from a run where half the
134
+ engines were missing is not the same answer as "no findings" from a complete one.
135
+
136
+ ### And publish the proof next to the diff
137
+
138
+ If your pipeline produced a fix record, hand it to the same action:
139
+
140
+ ```yaml
141
+ - uses: arthurpanhku/dvalincode@v0.18.0
142
+ with:
143
+ fix-record: fix-record.json
144
+ ```
145
+
146
+ The runner re-derives the record from the file alone โ€” recomputing its hash and
147
+ re-deriving its verdict from its own evidence โ€” and posts the result on the pull
148
+ request. A record that was edited after it was issued fails here, and fails the
149
+ job. The reviewer does not have to trust the pipeline that produced it, or us.
150
+
151
+ ```
152
+ ๐Ÿ” Verified Fix Record
153
+ โœ… ce504a995395 ยท VERIFIED ยท scan-and-checks
154
+ - repaired by claude-code โ€” recorded, and not consulted for this verdict
155
+ - targets: 1 before โ†’ 0 remaining
156
+ - coverage: complete โ†’ complete
157
+ - introduced: none (gate high/new)
158
+ - outcome: verified
159
+ - โœ“ test: `npm run test` (exit 0)
160
+ - audit chain: verify-eeb1bae7 @ 80881867270d
161
+ ```
162
+
163
+ A repair that regressed says so in the same place, and fails the job with it:
164
+
165
+ ```
166
+ โŒ 916e2eeaf065 ยท NOT VERIFIED ยท scan-and-checks
167
+ - introduced: 1 finding(s) the first scan did not report (gate high/new)
168
+ - critical dvalin/sql-injection โ€” src/db.ts:31
169
+ - outcome: regressed
170
+ ```
171
+
57
172
  ### Or let your agent call it
58
173
 
59
174
  If an agent is writing the code, something other than that agent has to check
@@ -63,15 +178,94 @@ it. DvalinCode is an MCP server, so any agent that speaks MCP can:
63
178
  claude mcp add dvalin -- npx -y dvalincode mcp-serve --workspace .
64
179
  ```
65
180
 
66
- `dvalin_scan` is read-only and deterministic โ€” no model runs, no credentials are
67
- needed, no files are touched โ€” so an agent can call it after every edit. A real
68
- call against a vulnerable fixture returns in ~170ms with a ~600 byte payload.
69
- The same server also exposes `dvalin_run_task` for delegating a whole governed
70
- task, plus the session and audit-evidence tools.
181
+ One command configures the editor you actually use:
182
+
183
+ ```sh
184
+ npx dvalincode mcp-install cursor # .cursor/mcp.json
185
+ npx dvalincode mcp-install vscode # .vscode/mcp.json
186
+ npx dvalincode mcp-install claude-code # .mcp.json
187
+ ```
188
+
189
+ The formats differ in a way that fails silently โ€” VS Code keys its servers under
190
+ `servers`, Cursor under `mcpServers` โ€” so the command writes the right one and
191
+ merges into whatever is already there. [Editors and MCP โ†’](integrations/mcp/)
192
+
193
+ `dvalin_scan` accepts `diff: "uncommitted"`, which reports only on what the
194
+ agent just wrote rather than everything the repository already carried โ€” the
195
+ difference between a usable answer and a wall of pre-existing findings. It never
196
+ runs a model, edits the target workspace, or persists Dvalin state, so clients
197
+ can allow the preview by default. When a finding will be repaired, the agent
198
+ explicitly calls `dvalin_begin_verification` to record a small local workflow;
199
+ it can then retrieve the finding by fingerprint and request an independent
200
+ re-scan through `dvalin_get_finding` and `dvalin_verify_findings`.
201
+
202
+ That last one is the point: an agent that has just written a repair can ask for
203
+ an independent verdict on it. Dvalin re-scans, runs the project's own checks
204
+ itself, and returns a **Verified Fix Record** โ€” what was targeted, what remains,
205
+ what the repair introduced that was not there before, the gate the verdict was
206
+ reached under, which commands ran and the exit codes Dvalin observed, and how
207
+ much of the codebase was actually covered. Whoever wrote the repair is recorded
208
+ and never consulted. `dvalin_verify_fix` re-derives such a record offline, so the reviewer
209
+ receiving it does not have to trust the tool that issued it.
210
+ [FVP-1 โ†’](docs/spec/FIX-VERIFICATION.md) Responses include MCP `structuredContent`; scanner
211
+ readiness is available through `dvalin_list_scanners`. The same server exposes
212
+ `dvalin_run_task` as an optional implementation helper, plus session and audit
213
+ evidence tools.
214
+
215
+ Every client below has been driven to a real tool call rather than only a
216
+ handshake โ€” which client, which version, and on what date is a table rather
217
+ than a sentence, because hand-written version numbers go stale quietly.
218
+ [Integration support โ†“](#-integration-support) ยท [Agent integrations โ†’](integrations/)
219
+
220
+ The repository also contains one [dual Codex/Claude plugin payload](integrations/dvalin-security/)
221
+ with both native manifests, the shared security-gate skill, and the local MCP
222
+ server configuration. Installing it makes the gate discoverable from task
223
+ context instead of requiring every developer to remember a scan prompt.
224
+ Codex honors the scan's read-only MCP annotation. Claude Code requires one
225
+ explicit MCP permission by design; the plugin documents the exact scan-only
226
+ allow rule instead of asking users to bypass all permissions.
227
+
228
+ ### Wherever you already work
229
+
230
+ One server, reached the way each tool expects:
231
+
232
+ | Harness | How Dvalin reaches it |
233
+ |---|---|
234
+ | **Claude Code** | [dual plugin](integrations/dvalin-security/) or `claude mcp add` ยท [standalone skill](integrations/claude-code/) |
235
+ | **Codex** | [dual plugin](integrations/dvalin-security/) or `codex mcp add` ยท [SARIF interop](integrations/codex-security/) with Codex Security |
236
+ | **Cursor** | `dvalincode mcp-install cursor` |
237
+ | **VS Code** | `dvalincode mcp-install vscode` ยท [extension](editors/vscode/) for Problems, coverage/gate status, and offline VFR verification โ€” *built, not yet published* |
238
+ | **Windsurf ยท Zed** | stdio MCP through their own settings โ€” [server command](integrations/mcp/) |
239
+ | **Any MCP client** | [MCP registry](https://registry.modelcontextprotocol.io/): `io.github.arthurpanhku/dvalincode` |
240
+ | **GitHub Actions** | [Marketplace action](https://github.com/marketplace/actions/dvalin-security-scan) โ€” findings inline on the pull request diff |
241
+ | **Any CI** | `dvalin scan . --fail-on high`, SARIF out for code scanning |
242
+
243
+ The MCP config formats are not interchangeable โ€” VS Code keys its servers under
244
+ `servers`, Cursor under `mcpServers`, and the wrong one fails silently โ€” so
245
+ `mcp-install` writes the right shape and merges into whatever is already there.
246
+ [Editors and MCP โ†’](integrations/mcp/)
247
+
248
+ ### Or interoperate with Codex Security
249
+
250
+ [Codex Security](https://github.com/openai/codex-security) can export a completed,
251
+ sealed scan as SARIF. Import that portable projection without coupling Dvalin to
252
+ Codex Security's private state directory:
253
+
254
+ ```sh
255
+ DVALIN_CODEX_SCAN_DIR=/tmp/codex-security-results
256
+ npx @openai/codex-security scan . --output-dir "$DVALIN_CODEX_SCAN_DIR"
257
+ npx @openai/codex-security export "$DVALIN_CODEX_SCAN_DIR" \
258
+ --export-format sarif --source-root "$PWD" --output /tmp/codex-security.sarif
259
+
260
+ dvalin import /tmp/codex-security.sarif .
261
+ dvalin scan . --fail-on high
262
+ ```
71
263
 
72
- Verified end to end with both Claude Code and Codex driving a real tool call
73
- against the published package, not only completing a handshake.
74
- [Agent integrations โ†’](integrations/)
264
+ The import creates stable Dvalin remediation cases; `--no-persist` validates the
265
+ handoff without changing the backlog. Keep Codex Security's original manifest,
266
+ findings, and coverage artifacts togetherโ€”Dvalin imports the SARIF finding
267
+ projection but does not rewrite its sealed bundle or reinterpret its coverage.
268
+ [Integration guide โ†’](integrations/codex-security/)
75
269
 
76
270
  ### Then let it fix what it found
77
271
 
@@ -132,6 +326,76 @@ dvalincode report verify # re-derive the hash chain of the last run's audit l
132
326
 
133
327
  ---
134
328
 
329
+ ## ๐Ÿงฉ Integration support
330
+
331
+ Code written with an AI assistant passes through four sets of hands before it
332
+ merges: the agent that writes it, the editor the developer reads it back in, the
333
+ pull request that gates it, and the reviewer who has to believe the result. A
334
+ security answer that exists in only one of those places is not a gate โ€” it is a
335
+ suggestion the next stage is free to ignore.
336
+
337
+ Dvalin is one MCP server and one deterministic scan behind all four, so the
338
+ answer does not change depending on who asks it.
339
+
340
+ | Stage of the loop | Where you are | How Dvalin gets there | Status |
341
+ |---|---|---|---|
342
+ | **Writing the code** | Claude Code | [dual plugin](integrations/dvalin-security/) ยท `claude mcp add` ยท `mcp-install claude-code` | โœ… session verified |
343
+ | | Codex | [dual plugin](integrations/dvalin-security/) ยท `codex mcp add` ยท [SARIF interop](integrations/codex-security/) | โœ… session verified ยท capture pending |
344
+ | | Cursor | `dvalincode mcp-install cursor` | โš™๏ธ config verified |
345
+ | | Windsurf ยท Zed | stdio MCP through their own settings | โš™๏ธ documented, unverified |
346
+ | | Any MCP client | registry `io.github.arthurpanhku/dvalincode` | โš™๏ธ published |
347
+ | **Reading it back** | VS Code | `mcp-install vscode` ยท [extension](editors/vscode/) for Problems, coverage and gate status | โœ… editor verified |
348
+ | **Gating the merge** | GitHub Actions | [Marketplace action](https://github.com/marketplace/actions/dvalin-security-scan) โ€” findings on the diff, fix records re-derived on the runner | โœ… runs on this repository's own CI |
349
+ | | Any CI | `dvalin scan . --fail-on high`, SARIF out for code scanning | โœ… the exit code is the contract |
350
+ | **Believing the result** | anyone, offline | `dvalin verify-fix record.json` | โœ… no workspace, no network, no Dvalin state |
351
+
352
+ **โœ…** means a real client was driven end to end and the tool call was observed.
353
+ **โš™๏ธ** means the configuration is generated and its shape is tested, but no
354
+ session has been captured. The difference is not smoothed over here, because a
355
+ config file that loads is not evidence that a tool was ever called.
356
+
357
+ ### What each claim rests on
358
+
359
+ | Client | Version | Checked | How |
360
+ |---|---|---|---|
361
+ | Claude Code | CLI 2.1.260 | 2026-09-04 | server connected, `dvalin_scan` called with only that tool allow-listed, three findings returned โ€” capture below |
362
+ | Claude Code | CLI 2.1.251 | 2026-08-31 | weekly [`harness-interop`](.github/workflows/harness-interop.yml) โ€” a real handshake against the built binary, which needs no credentials |
363
+ | Codex | CLI 0.149.1 | 2026-09-03 | by hand on macOS: scan called under `approval=never` in a read-only sandbox, no workflow state written |
364
+ | Codex | CLI 0.151.0 | 2026-08-31 | weekly `harness-interop` โ€” the server spec is accepted and stored as stdio. **Not a handshake**: the tool-call step stays skipped until `CODEX_API_KEY` is set |
365
+ | VS Code | 1.134.0 ยท extension 0.18.0 | 2026-09-03 | packaged VSIX in a clean profile; finding, gate and coverage rendered in the editor โ€” capture below |
366
+ | Cursor | โ€” | 2026-09-04 | `mcp-install cursor` writes `.cursor/mcp.json` under `mcpServers`; no session has been captured |
367
+
368
+ That weekly workflow exists because an earlier version of this claim named two
369
+ CLI versions by hand, both went stale within weeks, and nothing said so. It now
370
+ re-checks against whatever those tools shipped that week, so the dates above
371
+ either move on their own or stop moving in public.
372
+
373
+ ### The same finding, in each place
374
+
375
+ **Claude Code** โ€” one read-only tool allow-listed, and the MCP call itself:
376
+
377
+ ![A Claude Code session calling the Dvalin MCP server](docs/screenshots/10-claude-code-session.png)
378
+
379
+ **Codex** โ€” capture pending.
380
+
381
+ <!--
382
+ The Codex session capture belongs here, as docs/screenshots/11-codex-session.png.
383
+ To produce it on a machine with the Codex CLI installed and authenticated:
384
+
385
+ codex mcp add dvalin -- npx -y dvalincode mcp-serve --workspace .
386
+ codex exec --sandbox read-only \
387
+ "Scan this workspace with Dvalin and list every finding with rule id, file, line and severity."
388
+
389
+ Or set the CODEX_API_KEY repository secret, and .github/workflows/harness-interop.yml
390
+ stops skipping its "Codex calls a tool" step every Monday.
391
+ -->
392
+
393
+ **VS Code** โ€” the same scanner contract, as squiggles and a status bar:
394
+
395
+ ![A Dvalin finding and its coverage status inside VS Code](docs/screenshots/09-vscode-dvalin-integration.png)
396
+
397
+ ---
398
+
135
399
  ## ๐Ÿ›๏ธ And it survives a security review
136
400
 
137
401
  That last command is the part that matters once more than one person depends on
@@ -171,13 +435,14 @@ If you are the person who has to approve this class of tool, start at
171
435
 
172
436
  ## ๐ŸŽฏ Core Goal
173
437
 
174
- > **Make AI coding approvable for regulated and security-sensitive teams.**
438
+ > **Make every code-producing human or agent pass the same independent security gate.**
175
439
 
176
- DvalinCode is built as an **approvable agent runtime**, not just another coding
177
- agent app. The core product is not only "AI writes code"; it is the evidence a
178
- security, compliance, or platform team needs to safely allow AI coding in
179
- financial services, healthcare, internal enterprise platforms, and other
180
- confidential codebases.
440
+ DvalinCode is built as an **agent-compatible security runtime**, not another
441
+ general coding-agent benchmark entry. The core product is scan evidence,
442
+ policy, baseline, deterministic verification, and portable interfaces that a
443
+ human developer, an external agent, or CI can all call. The bundled coding agent
444
+ stays capable enough to implement and test focused remediations reliably; its
445
+ model prose never decides whether the security gate passed.
181
446
 
182
447
  - **Any model** โ€” every OpenAI-compatible endpoint is a first-class citizen, local models included. Your workflow should never be hostage to one vendor's pricing, rate limits, or quality swings.
183
448
  - **Safe by default** โ€” three-tier approvals with diff preview, an undo stack, and sandboxed shell execution. An agent you can trust on full-auto.
@@ -425,12 +690,12 @@ not claim third-party ISO certification.
425
690
  <img src="assets/hero.png" alt="Dvalin Security health showing 22/100 F with 4 high and 6 medium findings, and a Findings list locating each eval to a line in the NodeGoat-derived route" width="100%">
426
691
  </p>
427
692
 
428
- **The verified result โ€” the three `eval` call sites replaced by a constrained
429
- numeric parser, one new injection regression test, all four open-source engines
430
- complete, 0 findings, 100/100 ยท A:**
693
+ **The verified result โ€” a real model-driven Verify turn inspected the repair,
694
+ ran the regression tests and all four open-source engines, then the deterministic
695
+ server re-scan reported complete coverage, a passing gate, 0 findings, 100/100 ยท A:**
431
696
 
432
697
  <p align="center">
433
- <img src="assets/dvalin-scan-after.jpg" alt="Dvalin verified re-scan showing 100/100 A, zero findings, and the remediation run that produced it" width="100%">
698
+ <img src="assets/dvalin-verify-local.jpg" alt="A local Dvalin model-driven Verify run showing its test evidence beside a deterministic 100/100 A re-scan, complete four-engine coverage, and a passing gate" width="100%">
434
699
  </p>
435
700
 
436
701
  **Home โ†’ Code โ†’ Dvalin โ€” the current workspaces:**
@@ -496,6 +761,19 @@ Headless `run` and `mcp-serve` keep the same policy and audit chokepoint as
496
761
  the interactive clients. See the [unattended recipes](docs/RECIPES-UNATTENDED.md)
497
762
  for cron, CI, and external-agent examples.
498
763
 
764
+ A run that scanned or filed a fix record also carries a `verification` envelope
765
+ beside its answer โ€” in `json`, in `stream-json`, and in the MCP `dvalin_run_task`
766
+ result. `coverageStatus` is the *weakest* coverage the run has evidence of, so
767
+ one complete scan cannot speak for a partial one, and any record it produced is
768
+ listed by path for offline re-derivation. A CI gate can read it directly:
769
+
770
+ ```sh
771
+ jq -e '.verification.coverageStatus == "complete"' run.json || exit 1
772
+ ```
773
+
774
+ That is the same rule the pull-request comment applies, on the surface with no
775
+ human watching. [Harness mode โ†’](docs/HARNESS-MODE.md)
776
+
499
777
  ### Windows
500
778
 
501
779
  Download `dvalincode-v*-windows-x64.zip` from [Releases](https://github.com/arthurpanhku/dvalincode/releases/latest), unzip, then double-click `start.bat`.
@@ -671,7 +949,8 @@ RESTORE โ†’ COMPACT โ†’ COMMAND โ†’ BUILD โ†’ RUN โ†’ SAVE โ†’ RESPOND โ†’ DONE
671
949
  npm test
672
950
  ```
673
951
 
674
- **357 tests ยท 52 files ยท all green.**
952
+ **584 core tests ยท 74 files ยท all green.** The VS Code extension has a separate
953
+ 37-test suite plus one opt-in published-package integration test.
675
954
 
676
955
  ---
677
956
 
@@ -792,7 +1071,7 @@ Contributions welcome. The codebase is intentionally small and surgical โ€” see
792
1071
  ```sh
793
1072
  git clone https://github.com/arthurpanhku/dvalincode
794
1073
  cd dvalincode && npm install
795
- npm test # 357/357 โœ…
1074
+ npm test # 584/584 core tests โœ…
796
1075
  npm run typecheck
797
1076
  ```
798
1077
 
@@ -828,6 +1107,10 @@ ecosystem:
828
1107
  similar coding agents clarified user expectations around terminal agents,
829
1108
  plan/build modes, permission prompts, project-local context, sandboxing,
830
1109
  session lifecycle, MCP integration, and diff-first editing workflows.
1110
+ - [OpenAI Codex Security](https://github.com/openai/codex-security) and its
1111
+ public documentation informed Dvalin's specialist security-agent research and
1112
+ portable SARIF handoff. Dvalin remains an independent, competing security
1113
+ runtime.
831
1114
  - The `AGENTS.md` project-instruction convention, common in coding-agent tools,
832
1115
  informed DvalinCode's project-local instruction loading behavior.
833
1116
  - CodeQL, GitHub Code Scanning, Semgrep, SARIF, OpenSSF Scorecard, and ISO/IEC
@@ -845,31 +1128,22 @@ Full source references: [docs/REFERENCES.md](docs/REFERENCES.md)
845
1128
 
846
1129
  ---
847
1130
 
848
- ## โญ Star Growth
849
-
850
- <p align="center">
851
- <a href="https://www.star-history.com/#arthurpanhku/dvalincode&Date">
852
- <picture>
853
- <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=arthurpanhku/dvalincode&type=Date&theme=dark">
854
- <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/svg?repos=arthurpanhku/dvalincode&type=Date">
855
- <img alt="DvalinCode Star History Chart" src="https://api.star-history.com/svg?repos=arthurpanhku/dvalincode&type=Date">
856
- </picture>
857
- </a>
858
- </p>
859
-
860
- ---
861
-
862
1131
  ## ๐Ÿ’› Thanks to Our Contributors
863
1132
 
864
1133
  <p align="center">
865
1134
  Every issue, idea, documentation improvement, test, and code contribution helps make DvalinCode better.
866
1135
  </p>
867
1136
 
868
- <p align="center">
869
- <a href="https://github.com/arthurpanhku/dvalincode/graphs/contributors">
870
- <img src="https://contrib.rocks/image?repo=arthurpanhku/dvalincode" alt="DvalinCode contributors">
871
- </a>
872
- </p>
1137
+ | Contributor | GitHub profile |
1138
+ | --- | --- |
1139
+ | Arthur Pan | [@arthurpanhku](https://github.com/arthurpanhku) |
1140
+ | Shivas | [@shivasb42](https://github.com/shivasb42) |
1141
+ | Aditya | [@adity982](https://github.com/adity982) |
1142
+ | badhope | [@weed33834](https://github.com/weed33834) |
1143
+ | Samran Asif | [@webdevsamran](https://github.com/webdevsamran) |
1144
+ | dchaudhari7177 | [@dchaudhari7177](https://github.com/dchaudhari7177) |
1145
+
1146
+ See the [complete contribution history](https://github.com/arthurpanhku/dvalincode/graphs/contributors), including automated dependency and maintenance updates.
873
1147
 
874
1148
  <p align="center">
875
1149
  <sub>Want to join them? Read the <a href="CONTRIBUTING.md">contribution guide</a> and send your first pull request.</sub>