@agent-ix/quoin 0.5.3 → 0.5.5

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/README.md CHANGED
@@ -212,10 +212,50 @@ The technical specification for this library was itself authored with the `spec-
212
212
 
213
213
  ### Evals
214
214
 
215
- Eval scenarios live in [spec/evals.md](spec/evals.md) and run via the agent-pty-driven
216
- harness in `evals/` (`make evals` / `make evals-all`). The harness profiles a **real** agent
217
- running the skills/workflows and records token/tool/latency metrics from the Claude Code
218
- transcript. Unit tests cover the mechanical CLI behavior; see [evals/README.md](evals/README.md).
215
+ Eval scenarios live in [spec/evals.md](spec/evals.md) and run through the shared
216
+ [`@agent-ix/cli-agent-evals`](../cli-agent-evals) toolkit. The suite definition is
217
+ [`cli-agent-evals.config.mjs`](cli-agent-evals.config.mjs); quoin-specific fixtures,
218
+ module seeding, prompts, and assertions remain under [`evals/`](evals/).
219
+
220
+ Live runs profile a **real** agent running the skills/workflows and record token,
221
+ tool, and latency metrics from the available transcript. Unit tests cover the
222
+ mechanical CLI behavior.
223
+
224
+ ```bash
225
+ make evals
226
+ make evals-all
227
+ node ../cli-agent-evals/bin/cli-evals.js run \
228
+ --suite ./cli-agent-evals.config.mjs \
229
+ --canary \
230
+ --agent claude \
231
+ --model sonnet
232
+ ```
233
+
234
+ Agent plugin setup for authoring/running evals from an agent:
235
+
236
+ ```bash
237
+ claude plugin marketplace add agent-ix/cli-agent-evals
238
+ claude plugin install cli-agent-evals
239
+
240
+ codex plugin marketplace add agent-ix/cli-agent-evals
241
+ codex plugin add cli-agent-evals
242
+
243
+ gh skill install agent-ix/cli-agent-evals --all --scope user --agent opencode
244
+ gh skill install agent-ix/cli-agent-evals --all --scope user --agent github-copilot
245
+ ```
246
+
247
+ Minimal integration pattern:
248
+
249
+ ```js
250
+ import { defineSuite } from "../cli-agent-evals/dist/index.js";
251
+ import { SCENARIOS } from "./evals/scenarios/index.mjs";
252
+
253
+ export default defineSuite({
254
+ name: "quoin",
255
+ rootDir: import.meta.dirname,
256
+ scenarios: SCENARIOS,
257
+ });
258
+ ```
219
259
 
220
260
  ## About
221
261
 
@@ -349,7 +349,7 @@ function X(e) {
349
349
  return e || de();
350
350
  }
351
351
  function Z() {
352
- return X("0.5.3-dirty");
352
+ return X("0.5.5");
353
353
  }
354
354
  function fe(e) {
355
355
  return e.command === "update" ? le : e.command === "write" ? J : e.command === "catalog" ? se : e.command === "plugin" ? ce : R().includes(e.command) ? Y : q;
package/dist/cli.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"cli.d.ts","sourceRoot":"","sources":["../src/cli.ts"],"names":[],"mappings":"AAyMA,wBAAsB,IAAI,CAAC,IAAI,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,IAAI,CAAC,CA6CxD;AAqBD,wBAAgB,cAAc,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAGpD;AAED,wBAAgB,cAAc,IAAI,MAAM,CAEvC"}
1
+ {"version":3,"file":"cli.d.ts","sourceRoot":"","sources":["../src/cli.ts"],"names":[],"mappings":"AAyMA,wBAAsB,IAAI,CAAC,IAAI,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,IAAI,CAAC,CA6CxD;AAoBD,wBAAgB,cAAc,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAGpD;AAED,wBAAgB,cAAc,IAAI,MAAM,CAEvC"}
package/dist/cli.js CHANGED
@@ -1,2 +1,2 @@
1
- import { n as e, r as t, t as n } from "./cli-COr-fGKh.js";
1
+ import { n as e, r as t, t as n } from "./cli-DOC_L1Lg.js";
2
2
  export { n as main, e as packageVersion, t as resolveVersion };
package/dist/index.js CHANGED
@@ -1,2 +1,2 @@
1
- import { a as e, c as t, d as n, f as r, g as i, h as a, i as o, l as s, m as c, o as l, p as u, s as d, t as f, u as p } from "./cli-COr-fGKh.js";
1
+ import { a as e, c as t, d as n, f as r, g as i, h as a, i as o, l as s, m as c, o as l, p as u, s as d, t as f, u as p } from "./cli-DOC_L1Lg.js";
2
2
  export { o as createAuthoringPack, u as defaultModuleRoots, d as defaultModulesManifest, t as ensureDefaultModules, c as filamentModulesDir, a as findCatalogEntry, e as formatAuthoringPack, s as installPlugin, p as listPlugins, i as loadCatalog, f as main, n as parseSourceArg, l as parseTypeList, r as removePlugin };
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "@agent-ix/quoin",
3
3
  "description": "Spec-driven development for Claude Code — author validated, ISO/IEC/IEEE 29148-aligned specs, then plan and build against them.",
4
- "version": "0.5.3",
4
+ "version": "0.5.5",
5
5
  "license": "MIT",
6
6
  "type": "module",
7
7
  "files": [
@@ -50,8 +50,11 @@ All steps required!
50
50
  This skill produces or updates a **plan bundle** at `<project_root>/plan/<Plan-id>-<slug>/`:
51
51
 
52
52
  - `plan.md` (`type: Plan`): The central living document — requirements summary,
53
- dependency graph, execution tracks, quality gates, and test plan. Frontmatter
54
- `relationships:` trace (`references`) the requirements the plan covers.
53
+ dependency graph, execution tracks, quality gates, and test plan. It is an
54
+ **orchestration overview, not a mirror of the bundle**: ownership, the DAG, and test
55
+ traces are authoritative in the task frontmatter, so `plan.md` states each fact once
56
+ (see `references/step-3-execution-plan.md`). Frontmatter `relationships:` trace
57
+ (`references`) the requirements the plan covers.
55
58
  - `tasks/Task-NNN-<slug>.md` (`type: Task`): One file per task. Frontmatter carries
56
59
  the machine contract — `status`, `track`, `priority`, and `relationships:` edges
57
60
  (`depends_on` for the DAG, `references` for requirement ownership, `verifies` for
@@ -75,4 +75,14 @@ relationships:
75
75
  ### Cross-cutting constraints
76
76
  - `NFR-001` applies to [which code paths].
77
77
  - `NFR-002` applies to [which merge points].
78
+
79
+ ### The seams (optional)
80
+ One short paragraph grounding the plan in the existing code — the concrete files,
81
+ functions, or boundaries each track attaches to (e.g. "FR-021 widens the write guard in
82
+ `authorize.ts`; FR-022 reuses the existing `chokidar-change-source.ts`"). Keep it to a
83
+ few sentences; omit if the spec is greenfield.
78
84
  ```
85
+
86
+ This is the **single** dependency-graph representation for the bundle. Step 3 references
87
+ it — it does not re-emit a "Remaining Dependency Graph" or a separate "Execution Tracks"
88
+ section.
@@ -37,3 +37,7 @@
37
37
  ### Verification (NFRs)
38
38
  - [ ] **verify_response_time** (NFR-001): Run load test to ensure p95 < 200ms.
39
39
  ```
40
+
41
+ This is the **single** TC enumeration for the bundle. Downstream sections reference TC
42
+ ids only via the Step 3 **Task File Mapping** table — do not re-list them per track or in
43
+ track exit-criteria.
@@ -64,7 +64,8 @@ A gate is a checkpoint where you validate a fundamental assumption before invest
64
64
  - Before starting optimization work (confirm correctness first)
65
65
  - Before starting dependent work that would need rework if the gate fails
66
66
 
67
- **Gate format:**
67
+ **Gate format** (this is the thinking format; in `plan.md` the gate is rendered
68
+ **once**, as a one-liner inline in its track — see Output Format below):
68
69
  ```markdown
69
70
  #### Gate: [Name]
70
71
  - **Measures:** [What you're validating]
@@ -155,41 +156,46 @@ the `plan.md` body — see the Output Format below.
155
156
 
156
157
  ## Output Format (appended to plan/<Plan-id>-<slug>/plan.md)
157
158
 
159
+ `plan.md` is an **orchestration overview, not a mirror of the bundle.** Ownership, the
160
+ DAG, and test traces are authoritative in **task frontmatter** (`references`,
161
+ `depends_on`, `verifies`). State each fact **once**: the requirement checklist in
162
+ Requirements Summary (Step 1), the dependency *reasons* in the Dependency Graph (Step 1),
163
+ the TC enumeration in the Test Plan (Step 2), and the per-task row in the Task File
164
+ Mapping below. Step 3 appends only the sections below — it does **not** re-emit the
165
+ dependency graph, add a separate "Execution Tracks" section, restate any task's
166
+ `## Scope`, or re-list TC ids per track.
167
+
158
168
  ```markdown
159
169
  ## Remaining Work
160
170
 
161
- ### Remaining Dependency Graph
162
- [Text DAG or ASCII diagram showing what depends on what]
171
+ Describe the tracks **once**, here. One line per item: title, difficulty (optional
172
+ size estimate), and a *behavioral* exit criterion. The item's scope and subtasks live
173
+ in its task file — do not repeat them. Do not list TC ids here; they live in the Test
174
+ Plan and the Task File Mapping.
163
175
 
164
176
  ### Track A: Critical Path (serial)
165
- #### A1: [First step]
166
- - **Scope:** ...
167
- - **Difficulty:** Easy | Medium | Hard
168
- - **Estimated new code:** ~NNN lines
169
- - **Exit criteria:** ...
170
-
171
- #### A2: [Second step]
172
- ...
173
-
174
- #### Gate: [Quality checkpoint]
175
- - **Measures:** ...
176
- - **Pass criteria:** ...
177
+ - **A1 = Task-NNN** [title] — [Easy|Medium|Hard], ~NNN lines; exit: [behavioral outcome, e.g. "converges with no acknowledged loss"].
178
+ - **A2 = Task-NNN** [title] — [difficulty]; exit: [behavioral outcome].
179
+ - **Gate = Task-NNN** [name] — measures [what]; pass: [threshold]. (State each gate once, here in its track — never also in a separate section.)
177
180
 
178
181
  ### Track B: Parallel (independent agent, can start now)
179
- #### B1: [Parallel item]
180
- ...
182
+ - **B1 = Task-NNN** [title] — [difficulty]; exit: [behavioral outcome].
181
183
 
182
184
  ### Track C: Post-Gate
183
- #### C1: [Gated item]
184
- ...
185
+ - **C1 = Task-NNN** [title] — [difficulty]; exit: [behavioral outcome].
185
186
 
186
187
  ## Parallel Execution Summary
187
- [ASCII timeline showing tracks running concurrently]
188
+ [ASCII timeline showing the tracks running concurrently.]
188
189
 
189
190
  ## Task File Mapping
190
- [Table mapping Task-NNN files → track → owned requirements → status. The
191
- authoritative status/deps live in each task's frontmatter; this table is a
192
- human-readable mirror.]
191
+
192
+ The single per-task table — the only place that enumerates per-task ownership and
193
+ verification. It is a human-readable mirror of the task frontmatter, which stays
194
+ authoritative for status and edges.
195
+
196
+ | Task | Track | Owns (references) | Verified by (verifies) | Status |
197
+ | -------- | ----- | ----------------- | ---------------------- | ----------- |
198
+ | Task-NNN | A | FR-00X | TC-0XX…TC-0XX | not_started |
193
199
 
194
200
  ## Coordination Rules
195
201
  [What to freeze, what not to start early, shared-file/shared-state single-writer
@@ -198,6 +204,13 @@ discipline, merge sequencing.]
198
204
 
199
205
  ## Common Mistakes to Avoid
200
206
 
207
+ - **Mirroring frontmatter in prose.** `plan.md` is an orchestration overview; ownership,
208
+ the DAG, and test traces are authoritative in task frontmatter (`references`,
209
+ `depends_on`, `verifies`). State each fact once — requirement checklist (Requirements
210
+ Summary), dependency reasons (Dependency Graph), TC enumeration (Test Plan), per-task
211
+ row (Task File Mapping). Do not add a second "Execution Tracks" section, re-emit the
212
+ dependency graph, re-list TC ids in track exit-criteria, restate a task's `## Scope`,
213
+ or describe a gate twice.
201
214
  - **Bundling done + remaining work in one task.** If build is done but scan isn't, they're separate tasks even if they were originally planned together.
202
215
  - **Missing shared dependencies.** If two items need the same algorithm, extract it as a deliverable — don't let two parallel workers implement it independently.
203
216
  - **No gates.** If the whole system's value depends on one unproven property (e.g., recall quality), measure it before building everything around it.