@squadrules/mcp 5.13.6 → 5.13.8

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 (31) hide show
  1. package/dist/.tsbuildinfo +1 -1
  2. package/dist/cli/commands/export.d.ts +1 -0
  3. package/dist/cli/commands/export.d.ts.map +1 -1
  4. package/dist/cli/commands/export.js +23 -0
  5. package/dist/cli/commands/export.js.map +1 -1
  6. package/dist/embed-docs/mem/README.md +32 -14
  7. package/dist/embed-docs/mem/adapter-migration.md +59 -47
  8. package/dist/embed-docs/mem/author-adapter.md +112 -70
  9. package/dist/embed-docs/mem/bulk-insert-adapters-via-cli.md +59 -24
  10. package/dist/embed-docs/mem/challenge-type-guide.md +43 -13
  11. package/dist/embed-docs/mem/critic-loop-guide.md +38 -17
  12. package/dist/embed-docs/mem/critic-loop.md +24 -17
  13. package/dist/embed-docs/mem/protocol-linking-guide.md +10 -3
  14. package/dist/embed-docs/mem/refine-search.md +1 -1
  15. package/dist/resources/embedded-mcp-resources.js +12 -12
  16. package/dist/resources/embedded-mcp-resources.js.map +1 -1
  17. package/dist/tools/activate.d.ts.map +1 -1
  18. package/dist/tools/activate.js +7 -19
  19. package/dist/tools/activate.js.map +1 -1
  20. package/dist/tools/artifact-materialize-command.d.ts +22 -0
  21. package/dist/tools/artifact-materialize-command.d.ts.map +1 -0
  22. package/dist/tools/artifact-materialize-command.js +37 -0
  23. package/dist/tools/artifact-materialize-command.js.map +1 -0
  24. package/dist/tools/review-evidence-check.d.ts +2 -0
  25. package/dist/tools/review-evidence-check.d.ts.map +1 -1
  26. package/dist/tools/review-evidence-check.js +5 -3
  27. package/dist/tools/review-evidence-check.js.map +1 -1
  28. package/dist/ui/assets/{AccountPage-T4SMLR6H.js → AccountPage-Dj5D6Bhr.js} +1 -1
  29. package/dist/ui/assets/{index-C2EWn0mF.js → index-5bncLBJ8.js} +2 -2
  30. package/dist/ui/index.html +1 -1
  31. package/package.json +1 -1
@@ -1,5 +1,5 @@
1
1
  ---
2
- version: "5.13.6"
2
+ version: "5.13.8"
3
3
  slug: critic-loop
4
4
  title: Critic Loop
5
5
  ---
@@ -20,8 +20,11 @@ by an explicit iteration ceiling (see Verdict).
20
20
  ## Activation Patterns
21
21
 
22
22
  **Typically invoked by:** another protocol at the end of a plan, implement,
23
- validation, or review phase via `forward` with the stored adapter URI
24
- `squadrules://adapter/critic-loop`.
23
+ validation, or review phase, reached from a `subagent` contract whose
24
+ `subagent.mcp` names `forward` with the stored adapter URI
25
+ `squadrules://adapter/critic-loop`. The critic runs in the delegated context, not
26
+ on the audited agent's thread — that separation is why the type is `subagent` and
27
+ not `forward`.
25
28
 
26
29
  **Can be invoked directly when agent needs:**
27
30
  - "review phase output" / "verify plan" / "audit implementation"
@@ -35,6 +38,8 @@ validation, or review phase via `forward` with the stored adapter URI
35
38
  - Trust the executor's assertions without independent evidence.
36
39
  - Classify a claim as `verified` based only on absence of contradiction.
37
40
  - Keep investigating after a precondition failure that blocks verification.
41
+ - Re-delegate this review to another subagent: you are the separation the caller
42
+ asked for, and nesting a second layer of delegation hides the verdict's author.
38
43
 
39
44
  **Must Always:**
40
45
  - Extract claims before investigating.
@@ -43,6 +48,8 @@ validation, or review phase via `forward` with the stored adapter URI
43
48
  - Check every item in the supplied requirement list for coverage; a missing
44
49
  requirement is a FAIL even when the output is internally consistent.
45
50
  - Include confidence in every verdict and failure class in every FAIL verdict.
51
+ - Write `$VERDICT_FILE` on **every** path, including an early-abort, so the
52
+ caller's gate always reads a real verdict instead of failing on a missing file.
46
53
  - Preserve the boundary between execution and review.
47
54
 
48
55
  **Good trigger examples:**
@@ -117,14 +124,10 @@ are broken.
117
124
  - failure class = `verification_blocked`
118
125
  - confidence = `Low`
119
126
  - write focused diagnostics
120
- - skip all later steps
127
+ - write `$VERDICT_FILE` with those diagnostics, then skip all later steps
121
128
 
122
129
  **Output:** verified preconditions or early-abort diagnostics
123
130
 
124
- ```json
125
- {"contract":{"type":"shell","shell":{"cmd":"test -f \"${WORK_MANIFEST:-}\" || test -f \"${INPUT_FILE:-}\"","interpreter":"bash","timeout_seconds":3},"required":true}}
126
- ```
127
-
128
131
  ## Resolve Submission
129
132
 
130
133
  Turn the submission into the material under review. What cannot be resolved is
@@ -157,10 +160,6 @@ never guessed at.
157
160
 
158
161
  **Output:** resolved work inventory for the audit
159
162
 
160
- ```json
161
- {"contract":{"type":"subagent","subagent":{"context":"Resolve the work submission into reviewable material: read the manifest, resolve each item by kind from work.root, and record the resolved inventory with per-item status."},"required":true}}
162
- ```
163
-
164
163
  ## Audit
165
164
 
166
165
  Perform bounded adversarial review of the target against the resolved work
@@ -218,10 +217,6 @@ Any `missing` item — or any `partial` high-risk item — forces FAIL with
218
217
 
219
218
  **Output:** structured audit findings for the verdict step
220
219
 
221
- ```json
222
- {"contract":{"type":"subagent","subagent":{"context":"Perform bounded adversarial audit: extract claims, investigate with scope discipline, gather evidence by priority hierarchy, produce claim verdicts, review phase boundary, and check requirement coverage."},"required":true}}
223
- ```
224
-
225
220
  ## Verdict
226
221
 
227
222
  Turn the audit into a final PASS / FAIL result.
@@ -246,7 +241,8 @@ Turn the audit into a final PASS / FAIL result.
246
241
  - Line 1 is `PASS` → the loop is satisfied; the calling adapter may continue.
247
242
  - Line 1 is `FAIL` → the calling adapter **MUST NOT** proceed to reward, **MUST
248
243
  NOT** ask the user for approval, and **MUST NOT** continue. It must address the
249
- findings, increment `context.iteration`, and re-invoke critic-loop.
244
+ findings, increment `context.iteration`, and re-invoke critic-loop from a fresh
245
+ delegated subagent — never on the thread that produced the work under review.
250
246
  - `failure_class: iteration_ceiling` → stop looping and surface the unresolved
251
247
  findings; never silently pass. The ceiling bounds the loop so a weak agent
252
248
  cannot thrash forever.
@@ -265,3 +261,14 @@ The phase output has been independently challenged and looped until satisfied. T
265
261
  resulting verdict is more truthful because it is based on extracted claims,
266
262
  explicit evidence, contradiction search, requirement-by-requirement coverage, and
267
263
  a clean boundary between execution and review.
264
+
265
+ The single contract on this adapter is the verdict-file syntax gate: it is the
266
+ one artifact the critic produces, and everything before it is the critic's own
267
+ reasoning. The precondition, submission-resolution, and audit steps carry no
268
+ contracts because a contract gates an outcome, not a step, and a critic that
269
+ gates itself inward re-delegates work it was delegated to perform.
270
+
271
+ That gate asserts the verdict file's **shape**, not the review's soundness: a
272
+ `PASS` is this critic's judgment, and the separation that makes it worth
273
+ something is that a critic was run apart from the executor. The caller's own
274
+ `PASS` gate proves only that a file was written saying `PASS`.
@@ -1,5 +1,5 @@
1
1
  ---
2
- version: "5.13.6"
2
+ version: "5.13.8"
3
3
  slug: protocol-linking-guide
4
4
  title: Adapter Linking Guide
5
5
  ---
@@ -50,7 +50,12 @@ cohesive concern.
50
50
  - the adapter exceeds 350 lines
51
51
  - weaker models cannot follow the full adapter in one pass
52
52
  - runtime routing to different adapters is needed (Router pattern)
53
- - a step is reusable across multiple parent adapters (e.g. `critic-loop`)
53
+ - a step is reusable across multiple parent adapters (e.g. a shared
54
+ deployment-check adapter)
55
+
56
+ A review is **not** a chain link: reaching `critic-loop` to audit the current
57
+ run's work is a `subagent` hop, because a `forward` hop leaves the critic on the
58
+ thread that made the mistakes.
54
59
 
55
60
  Chaining is NOT a replacement for layers. A 6-step adapter that fits in 350
56
61
  lines should stay as one adapter with 6 H2 layers.
@@ -118,10 +123,12 @@ To run an extension adapter, call **`forward`** with
118
123
  Both system and user-authored adapters chain using slugs in `next_action`:
119
124
 
120
125
  ```text
121
- call forward with squadrules://adapter/critic-loop ...
122
126
  call forward with squadrules://adapter/implement-terraform ...
123
127
  ```
124
128
 
129
+ A review hop is not a chain link — reach `critic-loop` from a `subagent` contract
130
+ whose `subagent.mcp` names `forward`.
131
+
125
132
  Slugs are globally unique (enforced by `train`), so resolution is
126
133
  deterministic — no scoring, no ambiguity.
127
134
 
@@ -1,5 +1,5 @@
1
1
  ---
2
- version: "5.13.6"
2
+ version: "5.13.8"
3
3
  slug: refine-search
4
4
  title: Get help refining your search
5
5
  ---