@agent-ix/quoin 0.5.4 → 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 +44 -4
- package/dist/{cli-Dk-aYzJo.js → cli-DOC_L1Lg.js} +1 -1
- package/dist/cli.js +1 -1
- package/dist/index.js +1 -1
- package/package.json +1 -1
- package/skills/spec-to-plan/SKILL.md +5 -2
- package/skills/spec-to-plan/references/step-1-analysis.md +10 -0
- package/skills/spec-to-plan/references/step-2-test-plan.md +4 -0
- package/skills/spec-to-plan/references/step-3-execution-plan.md +36 -23
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
|
|
216
|
-
|
|
217
|
-
|
|
218
|
-
|
|
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.
|
|
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.js
CHANGED
|
@@ -1,2 +1,2 @@
|
|
|
1
|
-
import { n as e, r as t, t as n } from "./cli-
|
|
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-
|
|
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.
|
|
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.
|
|
54
|
-
|
|
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
|
-
|
|
162
|
-
|
|
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
|
-
|
|
166
|
-
- **
|
|
167
|
-
- **
|
|
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
|
-
|
|
180
|
-
...
|
|
182
|
+
- **B1 = Task-NNN** [title] — [difficulty]; exit: [behavioral outcome].
|
|
181
183
|
|
|
182
184
|
### Track C: Post-Gate
|
|
183
|
-
|
|
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
|
-
|
|
191
|
-
|
|
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.
|