empirical-sdd 0.22.0 → 0.24.0

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/CHANGELOG.md ADDED
@@ -0,0 +1,118 @@
1
+ # Changelog
2
+
3
+ All notable changes to Empirical SDD are documented in this file.
4
+
5
+ The format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
6
+ and this project follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html)
7
+ under the alpha rules in [docs/versioning.md](docs/versioning.md).
8
+
9
+ ## [Unreleased]
10
+
11
+ ## [0.24.0] - 2026-08-18
12
+
13
+ ### Added
14
+
15
+ - Added guided Linear, GitHub Projects, and Jira discovery/preview during Init,
16
+ provider-neutral semantic state suggestions, and strict equivalent MCP and
17
+ non-interactive setup surfaces.
18
+ - Added Tracker Policy v2 ticket behavior (`off`, `manual`, `ensure`), progress
19
+ visibility, automatic one-ticket reconciliation, idempotent milestone
20
+ comments, and receipt-approved evidence uploads or commit-pinned links.
21
+ - Added a packaged, runnable integration-repair demo that reproduces a
22
+ completed repository with missing activation artifacts and proves the
23
+ before, repair, and verified-after states without touching user data.
24
+
25
+ ### Changed
26
+
27
+ - Tracker synchronization now commits local state first, preserves user-authored
28
+ Linear descriptions, and resumes transition/comment/artifact effects from a
29
+ durable acknowledgement ledger.
30
+ - Existing Tracker Policy v1 repositories remain manual/legacy compatible and
31
+ repair preserves tracker bytes unless explicitly changed or disabled.
32
+ - Simplified the README around installation, everyday use, safety, and links to
33
+ the detailed project documentation.
34
+
35
+ ### Fixed
36
+
37
+ - Made Doctor validate both behavioral and non-behavioral integration receipts
38
+ without falsely treating a valid null capability claim as corruption.
39
+ - Made the shell-free GitHub delivery and publication runner reuse the host's
40
+ existing `gh` configuration locator without persisting credentials or secret
41
+ values in policy or runtime receipts.
42
+ - Made sanitized HTTPS pushes use the host's authenticated `gh` store through
43
+ an ephemeral, push-only Git helper configuration without inheriting `HOME`,
44
+ exporting tokens, or changing persistent Git configuration.
45
+ - Made Doctor report missing, drifted, or unsafe required project integrations
46
+ whenever Schema 5 setup is complete, instead of incorrectly reporting the
47
+ repository as activation-ready.
48
+ - Kept Doctor read-only and made its remediation point to explicit
49
+ `empirical-init`; repair recreates missing artifacts and updates
50
+ Empirical-owned content while preserving unmanaged conflicts for manual
51
+ resolution.
52
+ - Completed the lower stem of the terminal brand mark so its outline renders as
53
+ a closed cross.
54
+
55
+ ## [0.23.0] - 2026-08-11
56
+
57
+ Published through GitHub Actions trusted publishing with npm provenance.
58
+
59
+ ### Changed
60
+
61
+ - Replaced the globally automatic `empirical` skill with the explicit,
62
+ setup-only `empirical-init` bootstrap.
63
+ - Made initialized repositories route ordinary change prompts automatically
64
+ through marker-owned local instructions and skills; read-only prompts remain
65
+ outside the workflow.
66
+ - Added explicit-only invocation metadata where supported and retained all
67
+ existing workflow, evidence, tracker, integration, and publication gates.
68
+
69
+ ### Added
70
+
71
+ - Added this changelog and a documented alpha Semantic Versioning and release
72
+ policy.
73
+
74
+ ### Migration
75
+
76
+ - After upgrading from `0.22.x`, invoke `empirical-init` once in each existing
77
+ repository to install local automatic activation. Repair preserves Schema 5
78
+ configuration, context, feature history, and evidence unless setup values are
79
+ explicitly changed.
80
+
81
+ ## [0.22.0] - 2026-08-03
82
+
83
+ ### Added
84
+
85
+ - Introduced the Schema 5 protocol, strict Policy v2 evidence, resumable
86
+ journals, capability claims, independent integration, protected delivery,
87
+ explicit publication, and a single consolidated global workflow skill.
88
+
89
+ ## [0.20.4] - 2026-07-31
90
+
91
+ ### Fixed
92
+
93
+ - Completed the `0.20.4` release and its recorded release evidence.
94
+
95
+ ## [0.20.3] - 2026-07-31
96
+
97
+ ### Fixed
98
+
99
+ - Made CI and release fixtures portable across supported operating systems,
100
+ including Windows executable-extension casing.
101
+
102
+ ### Changed
103
+
104
+ - Simplified installation guidance and clarified README commands.
105
+
106
+ ## [0.20.2] - 2026-07-30
107
+
108
+ ### Changed
109
+
110
+ - Prepared and released package version `0.20.2`.
111
+
112
+ [Unreleased]: https://github.com/goempirical/empirical-sdd/compare/v0.24.0...HEAD
113
+ [0.24.0]: https://github.com/goempirical/empirical-sdd/compare/v0.23.0...v0.24.0
114
+ [0.23.0]: https://github.com/goempirical/empirical-sdd/compare/v0.22.0...v0.23.0
115
+ [0.22.0]: https://github.com/goempirical/empirical-sdd/compare/v0.20.4...v0.22.0
116
+ [0.20.4]: https://github.com/goempirical/empirical-sdd/compare/v0.20.3...v0.20.4
117
+ [0.20.3]: https://github.com/goempirical/empirical-sdd/compare/v0.20.2...v0.20.3
118
+ [0.20.2]: https://github.com/goempirical/empirical-sdd/releases/tag/v0.20.2
package/README.md CHANGED
@@ -1,120 +1,102 @@
1
1
  # Empirical SDD
2
2
 
3
- Agent-neutral, resumable spec-driven development with deterministic routing,
4
- immutable evidence, safe cross-worktree integration, and six agent-native
5
- skills. Empirical installs across 73 global agent targets and provides verified
6
- guidance for Codex, Claude Code, Cursor, Gemini CLI, Windsurf, and MCP clients.
3
+ Agent-neutral, resumable spec-driven development for coding agents. Empirical
4
+ turns an ordinary change request into a deterministic workflow with durable
5
+ state, reviewable evidence, and safe Git integration.
7
6
 
8
- > Empirical 0.22 is alpha software and intentionally introduces the breaking
9
- > Schema 5 protocol. One checkout selects at most one active feature; linked Git
10
- > worktrees isolate parallel work.
7
+ > Empirical 0.24 is alpha software. It requires Node.js 22 or newer.
11
8
 
12
9
  ## Install
13
10
 
11
+ ```sh
12
+ npm install -g empirical-sdd
13
+ empirical install
14
+ ```
15
+
16
+ Choose the agents you use, reload them, then run `empirical-init` once in each
17
+ repository. After initialization, ask for work normally—for example, “fix the
18
+ pagination bug.” Empirical activates automatically for change requests while
19
+ read-only questions stay outside the workflow.
20
+
21
+ Codex uses `$empirical-init`, Claude Code uses `/empirical-init`, and Windsurf
22
+ uses `@empirical-init`. Existing 0.22 repositories should invoke it once after
23
+ upgrading; configuration, history, and evidence are preserved.
24
+
25
+ Doctor verifies that completed repositories still have every required local
26
+ instruction, skill, and MCP bridge. If it reports missing or drifted project
27
+ integrations, invoke `empirical-init` explicitly to reconcile Empirical-owned
28
+ artifacts; unmanaged or unsafe conflicts are preserved and remain visible.
29
+
30
+ ## What it provides
31
+
32
+ - Deterministic Fast or Complex routing based on the request's risk.
33
+ - Resumable Specify, Design, Plan, Implement, Verify, Review, and Integrate phases.
34
+ - Immutable evidence tied to criteria, source state, and provenance.
35
+ - Isolated parallel work through linked Git worktrees.
36
+ - Guided Linear, GitHub Issues + Projects, or Jira ticket mirrors with automatic
37
+ binding, milestone comments, and safe evidence projection.
38
+ - Explicit, guarded delivery and npm publication boundaries.
39
+
40
+ Completion is reported only at the highest proven level: implemented,
41
+ verified, integrated, delivered, or published.
42
+
43
+ ## Tracker setup
44
+
45
+ `empirical-init` always shows a Tracker section. Keep Local-only to make no
46
+ provider requests, or select Linear, GitHub Projects, or Jira. Empirical reads
47
+ credential values only from the environment-variable names you provide,
48
+ discovers accessible targets and workflow states, proposes all seven semantic
49
+ mappings, and shows the complete secret-free policy before saving. Ambiguous
50
+ state suggestions require an explicit choice; simple boards may intentionally
51
+ reuse one provider state for several phases. Repair preserves an existing
52
+ tracker policy unless you explicitly change or disable it.
53
+
54
+ Tracker Policy v2 supports `off`, `manual`, and `ensure` ticket behavior plus
55
+ blockers/final, phase-milestone, or every-revision progress comments. `ensure`
56
+ validates a referenced ticket, reconciles the feature marker, and creates only
57
+ when neither identifies one unique ticket. Local workflow state commits first;
58
+ provider outages leave exact retry state and never rewind local work. Comments
59
+ preserve user-authored descriptions, and receipt-approved evidence is uploaded
60
+ or linked only after repository containment, media, size, and digest checks.
61
+ Existing Tracker Policy v1 files remain valid with manual binding and legacy
62
+ projection until deliberately upgraded.
63
+
64
+ ## CLI
65
+
14
66
  | Command | Purpose |
15
67
  | --- | --- |
16
- | `npm install -g empirical-sdd` | Install the CLI. Node.js 22 or newer is required. |
17
- | `empirical install` | Choose coding agents and install all six skills. |
18
- | `empirical update` | Upgrade Empirical and reconcile installed skills. |
19
- | `empirical uninstall` | Remove managed global skills, owned selection metadata, then the global package. |
20
-
21
- The installer uses a pinned local catalog. It remembers exact target IDs,
22
- deduplicates shared skill roots, and performs no runtime network fetch or `npx`
23
- execution. Automation can use repeatable `--agent`/`-a`, `--all`, `--yes`, and
24
- `--json` options.
68
+ | `empirical install` | Select agents and install or repair Empirical integrations. |
69
+ | `empirical update` | Upgrade the package and refresh installed integrations. |
70
+ | `empirical uninstall` | Remove Empirical-managed global files and the package. |
71
+ | `empirical --help` | Show commands and automation options. |
25
72
 
26
- Uninstall is fail-closed: interactive use shows the exact scope and defaults to
27
- cancel, while automation must use `empirical uninstall --yes` (optionally with
28
- `--json`). It removes only marker-owned global artifacts. Project `.empirical`
29
- history, evidence, and repository MCP/agent configuration are always preserved;
30
- unmanaged or unsafe global paths are also preserved and reported.
73
+ `empirical uninstall` preserves project `.empirical` history, evidence, and
74
+ repository configuration. Automation must confirm removal with `--yes`.
31
75
 
32
- ### Difference from the previous command surface
33
-
34
- | Previous 0.21 behavior | Updated behavior |
35
- | --- | --- |
36
- | Public lifecycle exposed Install and Update only. | Public lifecycle also exposes `empirical uninstall`. |
37
- | Removing six agent skills required manual per-agent cleanup. | One command scans every unique catalog root and removes only marker-owned skills. |
38
- | Selection metadata had to be found manually. | Valid Empirical-owned metadata is removed; invalid or user-owned metadata is preserved. |
39
- | Package removal was a separate manual npm command. | Confirmed uninstall runs exact `npm uninstall -g empirical-sdd` last. |
40
- | Project preservation was implicit. | Help, confirmation, human output, and JSON explicitly report preserved project state. |
76
+ ## Safety
41
77
 
42
- ## Skills
78
+ Empirical asks when a material product choice or permission is missing. It does
79
+ not bypass host permissions or branch protection, extract credentials,
80
+ force-write Git history, delete real worktrees, or infer publication. Releases
81
+ remain bound to an exact version, commit, tag, and npm dist-tag.
43
82
 
44
- These are coding-agent skills, not public shell workflow commands.
83
+ ## Documentation
45
84
 
46
- | Skill | Purpose |
47
- | --- | --- |
48
- | `empirical <request>` | Initialize, resume, route, and run in normal mode. |
49
- | `empirical-init` | Initialize or repair repository context without starting work. |
50
- | `empirical-spec <request>` | Draft a concrete Complex contract and stop for approval. |
51
- | `empirical-socratic <idea>` | Run the durable five-pass interview, then draft the contract. |
52
- | `empirical-loop` | Resume the selected feature from its exact current revision. |
53
- | `empirical-yolo <request>` | Run autonomously to an explicit safe completion ceiling. |
54
-
55
- Native invocation examples include `$empirical` in Codex, `/empirical` in
56
- Claude Code, and `@empirical` in Windsurf. Reload an agent after installation.
57
-
58
- ## Trust model
59
-
60
- Requests are classified into deterministic risk floors: contract-neutral,
61
- behavioral, sensitive, migration, integration, delivery, or publication. Fast
62
- is available only to contract-neutral work. Every higher floor uses Complex;
63
- wording a risky request as “quick” cannot demote it.
64
-
65
- Normal mode asks only when a material product choice or permission is missing.
66
- YOLO persists bounded standing authorization and asks only for genuine blockers
67
- before its authorized ceiling. It never bypasses host permissions or branch
68
- protection, force-writes Git, extracts credentials, deletes real worktrees or
69
- branches, replaces immutable releases, or infers publication.
70
-
71
- Fast is contract-neutral and ends at verified. Complex records an impact
72
- manifest and proceeds through Specify, Design, Plan, Implement, Verify, Review,
73
- and independent Integrate. An authorized delivery may continue through two
74
- protected GitHub PRs. Publication is always a separate explicit operation bound
75
- to an exact version, commit, tag, and dist-tag.
76
-
77
- Evidence is not a caller-supplied boolean. Empirical either executes a Policy v2
78
- command or collects a content-addressed artifact, then writes an immutable
79
- receipt tied to criteria, source state, and provenance. Completion reports only
80
- the highest proven level: implemented, verified, integrated, delivered, or
81
- published.
82
-
83
- ## Repository model
84
-
85
- Schema 5 stores strict Policy v2 configuration, Manifest v2 knowledge
86
- fingerprints, impact manifests, receipts, Git-common-dir capability claims, and
87
- hash-chained per-feature journals. Terminal journals compact transactionally to
88
- a verified snapshot boundary. `empirical_doctor` diagnoses schema, journal,
89
- lock, claim, toolchain, policy, knowledge, evidence, worktree, and delivery
90
- state without mutating the repository.
91
-
92
- Schema 4 repositories migrate atomically on the first mutating 0.22 operation.
93
- The migration validates a complete candidate tree before promotion and retains
94
- a recovery receipt. Earlier schemas must first be upgraded to Schema 4 with the
95
- version that created them.
85
+ [Protocol](docs/protocol.md) · [Architecture](docs/architecture.md) ·
86
+ [MCP and tracking](docs/mcp.md) · [Demo](docs/demo.md) ·
87
+ [Security](docs/security.md) · [Migration](docs/migration-v1.md) ·
88
+ [Versioning](docs/versioning.md) · [Changelog](CHANGELOG.md)
96
89
 
97
90
  ## Development
98
91
 
99
92
  Development requires Node.js 22+ and Bun. CI covers Node 22, 24, and 26.
100
93
 
101
- | Command | Purpose |
102
- | --- | --- |
103
- | `bun install` | Install dependencies. |
104
- | `bun run check` | Type-check the source. |
105
- | `bun run test` | Run the test suite. |
106
- | `bun run test:coverage` | Enforce aggregate and per-module coverage gates. |
107
- | `bun run test:dist` | Build and smoke-test the packaged CLI and MCP server. |
108
- | `bun run test:package` | Install and import the generated npm package. |
109
- | `bun run ci` | Run every required gate. |
110
-
111
- The package exposes only `.`, `./protocol`, `./mcp`, and `./integrations`.
94
+ ```sh
95
+ bun install
96
+ bun run ci
97
+ ```
112
98
 
113
- ## Documentation
114
-
115
- [Protocol](docs/protocol.md) · [Architecture](docs/architecture.md) ·
116
- [MCP](docs/mcp.md) · [Demo](docs/demo.md) · [Security](docs/security.md) ·
117
- [Migration](docs/migration-v1.md)
99
+ The package exports `.`, `./protocol`, `./mcp`, and `./integrations`.
118
100
 
119
101
  ## License
120
102
 
@@ -64,7 +64,7 @@ export declare const AGENT_SKILL_TARGETS: readonly [{
64
64
  readonly aliases: readonly ["claude"];
65
65
  readonly executables: readonly ["claude"];
66
66
  readonly invocation: "/empirical";
67
- readonly reload: "Restart Claude Code so it reloads the global Empirical skills, then invoke the desired /empirical skill.";
67
+ readonly reload: "Restart Claude Code so it reloads the global Empirical Init skill, then invoke /empirical-init for repository setup or repair.";
68
68
  readonly projectMcp: true;
69
69
  readonly handoff: true;
70
70
  }, {
@@ -104,7 +104,7 @@ export declare const AGENT_SKILL_TARGETS: readonly [{
104
104
  readonly detectPaths: readonly [".codex"];
105
105
  readonly executables: readonly ["codex"];
106
106
  readonly invocation: "$empirical";
107
- readonly reload: "Restart or reopen Codex so it rescans user skills, then invoke any installed $empirical skill.";
107
+ readonly reload: "Restart or reopen Codex so it rescans user skills, then invoke $empirical-init for repository setup or repair.";
108
108
  readonly projectMcp: true;
109
109
  readonly handoff: true;
110
110
  }, {
@@ -371,7 +371,7 @@ export declare const AGENT_SKILL_TARGETS: readonly [{
371
371
  readonly detectPaths: readonly [".codeium/windsurf"];
372
372
  readonly executables: readonly ["windsurf"];
373
373
  readonly invocation: "@empirical";
374
- readonly reload: "Reload Windsurf or start a new Cascade session, then invoke the desired @empirical skill.";
374
+ readonly reload: "Reload Windsurf or start a new Cascade session, then invoke @empirical-init for repository setup or repair.";
375
375
  readonly projectMcp: true;
376
376
  readonly handoff: true;
377
377
  }, {