@dotdotgod/claude-code 0.2.0 → 0.2.2
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 +14 -5
- package/commands/dd/load.md +4 -3
- package/package.json +1 -1
- package/skills/project-load/SKILL.md +4 -3
package/README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# @dotdotgod/claude-code
|
|
2
2
|
|
|
3
|
-
[](https://www.npmjs.com/package/@dotdotgod/claude-code) [](https://github.com/dotdotgod/dotdotgod-kit/tree/main/packages/claude-code) [](
|
|
3
|
+
[](https://www.npmjs.com/package/@dotdotgod/claude-code) [](https://github.com/dotdotgod/dotdotgod-kit/tree/main/packages/claude-code) [](https://github.com/dotdotgod/dotdotgod-kit/blob/main/LICENSE)
|
|
4
4
|
|
|
5
5
|
Claude Code adapter for dotdotgod's docs-first project-memory workflow.
|
|
6
6
|
|
|
@@ -8,7 +8,16 @@ Use this package when you want Claude Code to initialize shared project docs, lo
|
|
|
8
8
|
|
|
9
9
|
## Start Here
|
|
10
10
|
|
|
11
|
-
|
|
11
|
+
This package is a Claude Code plugin: a plugin manifest plus `/dd:*` commands and skills. Installing it from npm alone does not register the commands — Claude Code loads it through plugin registration or a local plugin directory.
|
|
12
|
+
|
|
13
|
+
Register the dotdotgod plugin marketplace inside Claude Code, then install the plugin:
|
|
14
|
+
|
|
15
|
+
```text
|
|
16
|
+
/plugin marketplace add dotdotgod/dotdotgod-kit
|
|
17
|
+
/plugin install dotdotgod@dotdotgod
|
|
18
|
+
```
|
|
19
|
+
|
|
20
|
+
Then use the bundled `/dd:*` commands in your repository:
|
|
12
21
|
|
|
13
22
|
```text
|
|
14
23
|
/dd:init
|
|
@@ -18,7 +27,7 @@ Install or register this package as a Claude Code plugin, then use the bundled `
|
|
|
18
27
|
/dd:impact
|
|
19
28
|
```
|
|
20
29
|
|
|
21
|
-
For local development from
|
|
30
|
+
For local development from a source checkout, load the plugin directory directly instead of installing:
|
|
22
31
|
|
|
23
32
|
```bash
|
|
24
33
|
claude --plugin-dir /path/to/dotdotgod/packages/claude-code
|
|
@@ -62,7 +71,7 @@ Projects can customize memory roles with `memory.areas`, and can customize trace
|
|
|
62
71
|
|
|
63
72
|
Claude Code can run local lifecycle hooks from Claude settings. dotdotgod does not require hooks: `/dd:init`, `/dd:load`, `/dd:plan`, `/dd:impact`, and the bundled skills work without them.
|
|
64
73
|
|
|
65
|
-
Use hooks only when you want opt-in reminders, validation, or local safety rails around the same loop: plan, implement, impact-review, verify, review, and archive. See [`hooks/README.md`](hooks/README.md) for current lifecycle notes and advisory examples.
|
|
74
|
+
Use hooks only when you want opt-in reminders, validation, or local safety rails around the same loop: plan, implement, impact-review, verify, review, and archive. See [`hooks/README.md`](https://github.com/dotdotgod/dotdotgod-kit/blob/main/packages/claude-code/hooks/README.md) for current lifecycle notes and advisory examples.
|
|
66
75
|
|
|
67
76
|
## Local Development
|
|
68
77
|
|
|
@@ -73,7 +82,7 @@ pnpm --filter @dotdotgod/claude-code run pack:dry-run
|
|
|
73
82
|
|
|
74
83
|
## Learn More
|
|
75
84
|
|
|
76
|
-
See the [root README](
|
|
85
|
+
See the [root README](https://github.com/dotdotgod/dotdotgod-kit#readme), [Context curation](https://github.com/dotdotgod/dotdotgod-kit/blob/main/docs/concept/CONTEXT_CURATION.md), [Context mechanics](https://github.com/dotdotgod/dotdotgod-kit/blob/main/docs/concept/CONTEXT_MECHANICS.md), [Memory area config](https://github.com/dotdotgod/dotdotgod-kit/blob/main/docs/spec/MEMORY_AREA_CONFIG.md), and [Traceability config](https://github.com/dotdotgod/dotdotgod-kit/blob/main/docs/spec/TRACEABILITY_CONFIG.md).
|
|
77
86
|
|
|
78
87
|
## Compared with Graphify-Style Memory
|
|
79
88
|
|
package/commands/dd/load.md
CHANGED
|
@@ -14,7 +14,7 @@ User focus, if provided: `$ARGUMENTS`
|
|
|
14
14
|
|
|
15
15
|
## Goal
|
|
16
16
|
|
|
17
|
-
Load the current repository's dotdotgod project memory in a read-only pass. Explicit manual loads should build the fuller curated project map. Prompt-injected refreshes and already-loaded sessions should
|
|
17
|
+
Load the current repository's dotdotgod project memory in a read-only pass. Explicit manual loads should build the fuller curated project map. Prompt-injected refreshes and already-loaded sessions should use the explicit compact load command and build a delta-oriented context map from shared agent instructions, README indexes, specs, architecture notes, test docs, active plans, and relevant archive notes. In Pi, `/dd:load` runs a full load and `/dd:load:compact` runs a compact load; free-form arguments are treated as focus text and never change the load mode.
|
|
18
18
|
|
|
19
19
|
Do not modify files during the load pass unless the user explicitly asks for edits after the summary.
|
|
20
20
|
|
|
@@ -26,9 +26,10 @@ Do not modify files during the load pass unless the user explicitly asks for edi
|
|
|
26
26
|
2. Prefer the bounded CLI snapshot when available.
|
|
27
27
|
- If the user prompt contains explicit project-memory refs such as `[[PLAN_MODE]]`, run `dotdotgod expand <root> "<prompt>" --json` before broad `grep` or `find` scans, then read the resolved candidates selectively.
|
|
28
28
|
- If the prompt has high-signal natural refs such as `PLAN_MODE`, `docs/spec/PLAN_MODE.md`, or quoted doc names, use `dotdotgod expand <root> "<prompt>" --fuzzy --json` before broad scans; avoid fuzzy expansion for low-signal generic words alone and respect configured fuzzy low-signal add/remove terms.
|
|
29
|
-
- If `dotdotgod` is installed or available in the repository, run `dotdotgod load-snapshot <root> --json`.
|
|
29
|
+
- If `dotdotgod` is installed or available in the repository, run `dotdotgod load-snapshot <root> --json`. This bounded CLI snapshot backs `/dd:load` and `/dd:load:compact`.
|
|
30
30
|
- If the local environment allows package execution but no `dotdotgod` binary is available, optionally run `npx @dotdotgod/cli load-snapshot <root> --json`.
|
|
31
|
-
- Treat the snapshot as the first-pass project-memory map for cache status, graph size, top memory areas, top related communities, and archive inclusion policy. Avoid expanding command/event-heavy details unless the user asks for a full or diagnostic load.
|
|
31
|
+
- Treat the snapshot as the first-pass project-memory map for cache trust/freshness status, graph size, memory policy, top memory areas, top related communities, and archive inclusion policy. Avoid expanding command/event-heavy details unless the user asks for a full or diagnostic load.
|
|
32
|
+
- Treat docs structure as metadata: use README.md paths as a book-like table of contents and expand a specific docs subtree only when the user request, a graph community, a memory area, or a README route identifies it.
|
|
32
33
|
- During load/planning, treat `dotdotgod status`, `load-snapshot`, `resolve`, `expand`, `graph impact`, `graph communities`, read-only `config`, and `index` as bounded context/status helpers. `load-snapshot`, `resolve`, `expand`, `graph impact`, `graph communities`, and `index` are project-content safe but may refresh ignored `.dotdotgod/` cache metadata. Avoid mutating scaffold/config commands such as `init` or `config init` unless the user explicitly asks for initialization or config creation.
|
|
33
34
|
- Use `dotdotgod graph impact <root> --changed <path> --yml` as a task-focused structured impact map when the user identifies a likely source/config/doc file. In Claude Code, `/dd:impact` can run the same review; in Codex, `dd:impact` is the command-like trigger for the `impact-review` skill.
|
|
34
35
|
- Use `grep` or `find` after `expand`, impact, and targeted reads when the task needs fallback discovery or raw source text search.
|
package/package.json
CHANGED
|
@@ -10,7 +10,7 @@ version: 1.0.0
|
|
|
10
10
|
|
|
11
11
|
## Goal
|
|
12
12
|
|
|
13
|
-
Load the current repository's dotdotgod project memory in a read-only pass. Explicit manual loads should build the fuller curated project map. Prompt-injected refreshes and already-loaded sessions should
|
|
13
|
+
Load the current repository's dotdotgod project memory in a read-only pass. Explicit manual loads should build the fuller curated project map. Prompt-injected refreshes and already-loaded sessions should use the explicit compact load command and build a delta-oriented context map from shared agent instructions, README indexes, specs, architecture notes, test docs, active plans, and relevant archive notes. In Pi, `/dd:load` runs a full load and `/dd:load:compact` runs a compact load; free-form arguments are treated as focus text and never change the load mode.
|
|
14
14
|
|
|
15
15
|
Do not modify files during the load pass unless the user explicitly asks for edits after the summary.
|
|
16
16
|
|
|
@@ -22,9 +22,10 @@ Do not modify files during the load pass unless the user explicitly asks for edi
|
|
|
22
22
|
2. Prefer the bounded CLI snapshot when available.
|
|
23
23
|
- If the user prompt contains explicit project-memory refs such as `[[PLAN_MODE]]`, run `dotdotgod expand <root> "<prompt>" --json` before broad `grep` or `find` scans, then read the resolved candidates selectively.
|
|
24
24
|
- If the prompt has high-signal natural refs such as `PLAN_MODE`, `docs/spec/PLAN_MODE.md`, or quoted doc names, use `dotdotgod expand <root> "<prompt>" --fuzzy --json` before broad scans; avoid fuzzy expansion for low-signal generic words alone and respect configured fuzzy low-signal add/remove terms.
|
|
25
|
-
- If `dotdotgod` is installed or available in the repository, run `dotdotgod load-snapshot <root> --json`.
|
|
25
|
+
- If `dotdotgod` is installed or available in the repository, run `dotdotgod load-snapshot <root> --json`. This bounded CLI snapshot backs `/dd:load` and `/dd:load:compact`.
|
|
26
26
|
- If the local environment allows package execution but no `dotdotgod` binary is available, optionally run `npx @dotdotgod/cli load-snapshot <root> --json`.
|
|
27
|
-
- Treat the snapshot as the first-pass project-memory map for cache status, graph size, top memory areas, top related communities, and archive inclusion policy. Avoid expanding command/event-heavy details unless the user asks for a full or diagnostic load.
|
|
27
|
+
- Treat the snapshot as the first-pass project-memory map for cache trust/freshness status, graph size, memory policy, top memory areas, top related communities, and archive inclusion policy. Avoid expanding command/event-heavy details unless the user asks for a full or diagnostic load.
|
|
28
|
+
- Treat docs structure as metadata: use README.md paths as a book-like table of contents and expand a specific docs subtree only when the user request, a graph community, a memory area, or a README route identifies it.
|
|
28
29
|
- During load/planning, treat `dotdotgod status`, `load-snapshot`, `resolve`, `expand`, `graph impact`, `graph communities`, read-only `config`, and `index` as bounded context/status helpers. `load-snapshot`, `resolve`, `expand`, `graph impact`, `graph communities`, and `index` are project-content safe but may refresh ignored `.dotdotgod/` cache metadata. Avoid mutating scaffold/config commands such as `init` or `config init` unless the user explicitly asks for initialization or config creation.
|
|
29
30
|
- Use `dotdotgod graph impact <root> --changed <path> --yml` as a task-focused structured impact map when the user identifies a likely source/config/doc file. In Claude Code, `/dd:impact` can run the same review; in Codex, `dd:impact` is the command-like trigger for the `impact-review` skill.
|
|
30
31
|
- Use `grep` or `find` after `expand`, impact, and targeted reads when the task needs fallback discovery or raw source text search.
|