@orderful/droid 0.44.1 → 0.45.1

package/dist/tools/code-review/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid-code-review",
-  "version": "0.2.3",
+  "version": "0.2.4",
   "description": "Comprehensive code review using specialized agents. Reviews PRs, staged changes, branches, or specific files with confidence scoring.",
   "author": {
     "name": "Orderful",

package/dist/tools/code-review/TOOL.yaml

@@ -1,6 +1,6 @@
 name: code-review
 description: "Comprehensive code review using specialized agents. Reviews PRs, staged changes, branches, or specific files with confidence scoring."
-version: 0.2.3
+version: 0.2.4
 status: alpha
 
 includes:

package/dist/tools/code-review/skills/code-review/SKILL.md

@@ -18,6 +18,12 @@ Code-review has no configuration of its own. Optional integration with other too
 
 **Overrides:** This skill supports user-defined overrides. See `/droid` skill § Skill Overrides.
 
+## Custom Instructions
+
+Any command accepts a ` -- {instruction}` suffix. Split on the **first** ` -- ` (space-dash-dash-space): left is the command and its args, right is additional context or instruction to carry through the entire execution.
+
+Example: `/code-review #123 -- pay close attention to SQL injection risks`
+
 ## How It Works
 
 The `/code-review` command orchestrates multiple specialized agents in parallel:

package/dist/tools/codex/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid-codex",
-  "version": "0.3.0",
+  "version": "0.3.1",
   "description": "Shared organizational knowledge - PRDs, tech designs, domains, proposals, patterns, and explored topics. Self-describing: structure and workflows defined in codex repo. Use when loading project context, searching codex, capturing decisions, or creating new entries.",
   "author": {
     "name": "Orderful",

package/dist/tools/codex/TOOL.yaml

@@ -1,6 +1,6 @@
 name: codex
 description: "Shared organizational knowledge - PRDs, tech designs, domains, proposals, patterns, and explored topics. Self-describing: structure and workflows defined in codex repo. Use when loading project context, searching codex, capturing decisions, or creating new entries."
-version: 0.3.0
+version: 0.3.1
 status: beta
 
 includes:

package/dist/tools/codex/skills/codex/SKILL.md

@@ -86,6 +86,12 @@ If prerequisites fail, guide user to fix:
 
 **Overrides:** This skill supports user-defined overrides. See `/droid` skill § Skill Overrides.
 
+## Custom Instructions
+
+Any command accepts a ` -- {instruction}` suffix. Split on the **first** ` -- ` (space-dash-dash-space): left is the command and its args, right is additional context or instruction to carry through the entire execution.
+
+Example: `/codex search webhook -- focus on the retry logic patterns`
+
 ## Commands
 
 All codex operations follow workflows defined in the codex repo's `.codex/workflows/` folder.

package/dist/tools/comments/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid-comments",
-  "version": "0.3.5",
+  "version": "0.3.6",
   "description": "Enable inline conversations using @droid/@user markers. Tag @droid to ask the AI, AI responds with @{your-name}. Use /comments check to address markers, /comments cleanup to remove resolved threads. Ideal for code review notes and async collaboration.",
   "author": {
     "name": "Orderful",

package/dist/tools/comments/TOOL.yaml

@@ -1,6 +1,6 @@
 name: comments
 description: "Enable inline conversations using @droid/@user markers. Tag @droid to ask the AI, AI responds with @{your-name}. Use /comments check to address markers, /comments cleanup to remove resolved threads. Ideal for code review notes and async collaboration."
-version: 0.3.5
+version: 0.3.6
 status: beta
 
 includes:

package/dist/tools/comments/skills/comments/SKILL.md

@@ -63,6 +63,12 @@ Think of it like addressing someone in conversation:
 
 Inline comments shine for localized questions and actions within a file. For longer back-and-forth discussions in documents, they work great too - just be mindful that very long threads may be better moved to dedicated docs.
 
+## Custom Instructions
+
+Any command accepts a ` -- {instruction}` suffix. Split on the **first** ` -- ` (space-dash-dash-space): left is the command and its args, right is additional context or instruction to carry through the entire execution.
+
+Example: `/comments check -- keep responses brief, we're in a tight review cycle`
+
 ## Context Gathering
 
 When you find a `@droid` marker:

package/dist/tools/droid/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid",
-  "version": "0.7.0",
+  "version": "0.7.1",
   "description": "Core droid meta-skill for update awareness, tool discovery, and usage help. Checks for updates, helps users find tools, and answers 'how do I...' questions about droid workflows.",
   "author": {
     "name": "Orderful",

package/dist/tools/droid/TOOL.yaml

@@ -1,6 +1,6 @@
 name: droid
 description: "Core droid meta-skill for update awareness, tool discovery, and usage help. Checks for updates, helps users find tools, and answers 'how do I...' questions about droid workflows."
-version: 0.7.0
+version: 0.7.1
 status: beta
 
 # System tool - always stays current regardless of auto-update settings

package/dist/tools/droid/skills/droid/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: droid
 description: "Core droid meta-skill for updates, tool discovery, usage help, and skill overrides. Use when checking for droid updates, discovering available tools, answering 'how do I...' questions about droid workflows, or customizing skill behaviour. User prompts like 'any droid updates?', 'what tools do I have?', 'what tools do you have?', 'remind me what tools you have', 'hey droid what can you do', 'how do I add context?', 'remind me how to use codex', 'what's the best way to start a session?', 'customize brain search', 'create an override for brain'."
+alwaysApply: true
 allowed-tools: [Bash, Read, Write]
 ---
 

package/dist/tools/droid/skills/droid-bootstrap/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: droid-bootstrap
 description: "Lightweight droid bootstrap for update awareness. Checks for droid updates once per session."
+alwaysApply: true
 allowed-tools: [Bash]
 ---
 

package/dist/tools/edi-schema/.claude-plugin/plugin.json

@@ -0,0 +1,25 @@
+{
+  "name": "droid-edi-schema",
+  "version": "0.1.0",
+  "description": "Query EDI schemas without context overhead. Uses a sub-agent to inspect large schema and code-list files, then returns concise results.",
+  "author": {
+    "name": "Orderful",
+    "url": "https://github.com/orderful"
+  },
+  "repository": "https://github.com/orderful/droid",
+  "license": "MIT",
+  "keywords": [
+    "droid",
+    "ai",
+    "edi-schema"
+  ],
+  "skills": [
+    "./skills/edi-schema/SKILL.md"
+  ],
+  "commands": [
+    "./commands/edi-schema.md"
+  ],
+  "agents": [
+    "./agents/edi-schema-agent.md"
+  ]
+}

package/dist/tools/edi-schema/TOOL.yaml

@@ -0,0 +1,29 @@
+name: edi-schema
+description: "Query EDI schemas without context overhead. Uses a sub-agent to inspect large schema and code-list files, then returns concise results."
+version: 0.1.0
+status: beta
+
+includes:
+  skills:
+    - name: edi-schema
+      required: true
+  commands:
+    - name: edi-schema
+      is_alias: false
+  agents:
+    - edi-schema-agent
+
+dependencies: []
+
+prerequisites:
+  - name: jq
+    description: JSON query tool used for targeted schema extraction
+    check: "jq --version"
+    install_hint: "brew install jq"
+
+config_schema:
+  schemas_dir:
+    type: string
+    description: "Optional absolute path to a repository containing libs/schemas/. If empty, auto-detect from current workspace."
+    required: false
+    default: ""

package/dist/tools/edi-schema/agents/edi-schema-agent.md

@@ -0,0 +1,97 @@
+---
+name: edi-schema-agent
+description: "Specialized EDI schema query agent. Uses jq and targeted reads to extract precise schema details with minimal output."
+tools:
+  - Read
+  - Glob
+  - Grep
+  - Bash
+color: blue
+---
+
+You are an EDI schema analysis agent. Your job is to query large schema files and return concise, accurate findings.
+
+## Inputs
+
+You will receive:
+
+1. `schemasRoot`: Absolute repo path containing `libs/schemas/src/lib/data`
+2. `mode`: One of `field-search`, `segments`, `loops`, `codes`, `versions`, `types`
+3. `transactionType`: Full or short transaction type identifier
+4. `schemaFile` (optional): Resolved schema file path for schema-specific modes
+5. `query` (optional): Search text for `field-search`
+6. `elementId` (optional): Element identifier for `codes`
+
+## Rules
+
+- Prioritize correctness over broad output.
+- Never output full JSON blobs.
+- Keep returned content compact and actionable.
+- Prefer jq for extraction. If jq is missing, use `rg` + targeted reads.
+
+## Query Strategies
+
+### `field-search`
+
+Run exact identifier and title-first queries before fuzzy search:
+
+```bash
+jq '.. | objects | select(.elementIdentifier? == $id)' --arg id "{query}" "{schemaFile}"
+jq '.. | objects | select(.title? == $title)' --arg title "{query}" "{schemaFile}"
+jq '.. | objects | select(.title? | test($q; "i"))' --arg q "{query}" "{schemaFile}"
+```
+
+Return top matches including:
+
+- `elementIdentifier` (if present)
+- `title`
+- `schemaType`
+- `X12Requirement` / relevant requirement field
+- a concise path hint when derivable
+
+### `segments`
+
+```bash
+jq '[.. | objects | select(.schemaType? == "segmentInstance") | {title, requirement: .X12Requirement}] | unique' "{schemaFile}"
+```
+
+Return deduplicated segment names and requirement info.
+
+### `loops`
+
+```bash
+jq '[.. | objects | select(.schemaType? | test("loop")) | {title, schemaType}]' "{schemaFile}"
+```
+
+Return loop hierarchy summary. Include parent-child nesting hints if visible.
+
+### `codes`
+
+Find likely code-list file (for example `x12_004010_codeList.json`), then:
+
+```bash
+jq --arg id "{elementId}" '.[$id]' "{codeListFile}"
+```
+
+Return code values and descriptions. If no entry exists, return "not found" and suggest checking identifier/version.
+
+### `versions`
+
+List available schema filenames matching type in x12 + edifact directories, then normalize to type/version list.
+
+### `types`
+
+Read `transactionTypes.json` and return a concise sorted list (or filtered list if query provided).
+
+## Output Format
+
+Respond in markdown with this structure:
+
+1. `Resolved:` line (type/version and file used)
+2. `Results:` compact bullets or table
+3. `Notes:` any ambiguity, fallbacks, or missing data
+
+If no match:
+
+- state that clearly
+- provide 2-5 nearest alternatives when possible

package/dist/tools/edi-schema/commands/edi-schema.md

@@ -0,0 +1,33 @@
+---
+name: edi-schema
+description: "Query EDI schemas (X12/EDIFACT) without context blowup. Find fields, segments, loops, and code-list values."
+argument-hint: "{transactionType|types|versions {transactionType}} [{query|--segments|--loops|--codes {elementId}}]"
+---
+
+# /edi-schema
+
+**User invoked:** `/edi-schema $ARGUMENTS`
+
+**Your task:** Invoke the **edi-schema skill** with these arguments.
+
+## Examples
+
+- `/edi-schema 944_004010 freeFormDescription`
+- `/edi-schema 850 --segments`
+- `/edi-schema 856_004010 --loops`
+- `/edi-schema 850 --codes BEG01`
+- `/edi-schema types`
+- `/edi-schema versions 850`
+
+## Quick Reference
+
+```bash
+/edi-schema {transactionType} {query}     # Search fields/elements
+/edi-schema {transactionType} --segments  # List segment set
+/edi-schema {transactionType} --loops     # Show loop structure
+/edi-schema {transactionType} --codes ID  # Code-list values for element
+/edi-schema types                         # List transaction types
+/edi-schema versions {transactionType}    # List available versions
+```
+
+See the **edi-schema skill** for full behaviour and resolution rules.

package/dist/tools/edi-schema/skills/edi-schema/SKILL.md

@@ -0,0 +1,86 @@
+---
+name: edi-schema
+description: "Query EDI schemas without consuming parent context. Use when finding fields, loops, segments, requirements, or code-list values for X12/EDIFACT transaction types. User prompts like '/edi-schema 944_004010 freeFormDescription' or '/edi-schema 850 --codes BEG01'."
+argument-hint: "{transactionType|types|versions {transactionType}} [{query|--segments|--loops|--codes {elementId}}]"
+allowed-tools: [Read, Glob, Grep, Bash, Task]
+---
+
+# EDI Schema Skill
+
+Query EDI schemas through a sub-agent so large schema JSON files never enter parent context.
+
+## Commands
+
+```bash
+/edi-schema {transactionType} {query}
+/edi-schema {transactionType} --segments
+/edi-schema {transactionType} --loops
+/edi-schema {transactionType} --codes {elementId}
+/edi-schema types
+/edi-schema versions {transactionType}
+```
+
+## Input Rules
+
+- `transactionType` accepts either full form (`850_004010`, `ORDERS_D96A`) or short form (`850`, `ORDERS`).
+- If short form is used, resolve to the most common version and state which version was selected.
+- If multiple plausible versions exist and no safe default is clear, show options and ask user to pick.
+
+## Data Sources
+
+Expected paths in the target schemas repo:
+
+- `libs/schemas/src/lib/data/transactionSchemas/x12/`
+- `libs/schemas/src/lib/data/transactionSchemas/edifact/`
+- `libs/schemas/src/lib/data/codeLists/`
+- `libs/schemas/src/lib/data/transactionTypes.json`
+
+## Procedure
+
+### 1. Resolve schemas root
+
+1. Check `droid config --get tools.edi-schema.schemas_dir`.
+2. If configured and valid, use it.
+3. Otherwise auto-detect from current workspace by finding `libs/schemas/src/lib/data/`.
+4. If not found, ask user for the repo path or suggest setting `tools.edi-schema.schemas_dir`.
+
+### 2. Route command
+
+- `types`:
+  - List transaction types from `transactionTypes.json` (type + description where available).
+- `versions {transactionType}`:
+  - List matching schema versions from x12/edifact schema filenames.
+- `{transactionType} --segments`:
+  - Ask sub-agent for unique segment list and requirement indicators.
+- `{transactionType} --loops`:
+  - Ask sub-agent for loop hierarchy.
+- `{transactionType} --codes {elementId}`:
+  - Ask sub-agent for code-list values for the requested element ID.
+- `{transactionType} {query}`:
+  - Ask sub-agent to search by identifier/title/business term and return best matches.
+
+### 3. Sub-agent handoff
+
+Use `edi-schema-agent` for all schema/content queries. Pass:
+
+- Resolved schemas root
+- Resolved schema file path + transaction type/version
+- Query mode (`field-search`, `segments`, `loops`, `codes`, `versions`, `types`)
+- User input (`query`, `elementId`)
+
+### 4. Response format
+
+Keep responses compact and structured:
+
+- Header with resolved schema (`850_004010`, etc.)
+- Result table or bullets with identifiers and titles
+- Requirement/type constraints where relevant
+- Path hints (loop/segment path) when available
+- Clear next step if no matches
+
+## Behaviour Rules
+
+- Never dump the entire schema JSON.
+- Prefer exact identifier match first (for example `BEG01`) before fuzzy title matching.
+- When no exact match, return top fuzzy matches with confidence notes.
+- If jq is unavailable, fall back to `rg`/targeted reads and state that fallback was used.

package/dist/tools/meeting/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid-meeting",
-  "version": "0.1.1",
+  "version": "0.1.2",
   "description": "Work with meeting notes, summaries, and transcripts. List recent meetings, search content, generate context-aware summaries, export to codex. Backed by Granola MCP.",
   "author": {
     "name": "Orderful",

package/dist/tools/meeting/TOOL.yaml

@@ -1,6 +1,6 @@
 name: meeting
 description: "Work with meeting notes, summaries, and transcripts. List recent meetings, search content, generate context-aware summaries, export to codex. Backed by Granola MCP."
-version: 0.1.1
+version: 0.1.2
 status: beta
 
 includes:

package/dist/tools/meeting/skills/meeting/SKILL.md

@@ -35,6 +35,12 @@ If not configured, tell user:
 
 If connected, proceed.
 
+## Custom Instructions
+
+Any command accepts a ` -- {instruction}` suffix. Split on the **first** ` -- ` (space-dash-dash-space): left is the command and its args, right is additional context or instruction to carry through the entire execution. Note: flag-style `--all` uses `--flag` syntax (no surrounding spaces) and is distinct from this separator.
+
+Example: `/meeting search partner testing review -- extract action items only`
+
 ## Commands
 
 | Command | Action |

package/dist/tools/plan/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid-plan",
-  "version": "0.1.4",
+  "version": "0.1.5",
   "description": "Task-scoped planning with portable, structured plans. Use when planning implementation for a PR, ticket, or small feature. User prompts like 'let's plan this', 'can we start a plan', 'think through the implementation'.",
   "author": {
     "name": "Orderful",

package/dist/tools/plan/TOOL.yaml

@@ -1,6 +1,6 @@
 name: plan
 description: "Task-scoped planning with portable, structured plans. Use when planning implementation for a PR, ticket, or small feature. User prompts like 'let's plan this', 'can we start a plan', 'think through the implementation'."
-version: 0.1.4
+version: 0.1.5
 status: alpha
 
 includes:

package/dist/tools/plan/skills/plan/SKILL.md

@@ -30,6 +30,12 @@ Uses config from dependencies:
 
 **Overrides:** This skill supports user-defined overrides. See `/droid` skill § Skill Overrides.
 
+## Custom Instructions
+
+Any command accepts a ` -- {instruction}` suffix. Split on the **first** ` -- ` (space-dash-dash-space): left is the command and its args, right is additional context or instruction to carry through the entire execution.
+
+Example: `/plan new auth-refactor -- we must stay backward compatible with v1 clients`
+
 ## Commands
 
 | Command                | Action                                              |

package/dist/tools/project/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid-project",
-  "version": "0.3.4",
+  "version": "0.3.6",
   "description": "Manage project context files for persistent AI memory across sessions. Load, update, or create project context before working on multi-session features.",
   "author": {
     "name": "Orderful",

package/dist/tools/project/TOOL.yaml

@@ -1,6 +1,6 @@
 name: project
 description: "Manage project context files for persistent AI memory across sessions. Load, update, or create project context before working on multi-session features."
-version: 0.3.4
+version: 0.3.6
 status: beta
 
 includes:

package/dist/tools/project/skills/project/SKILL.md

@@ -36,6 +36,12 @@ Chat history disappears. Projects persist.
 
 **Overrides:** This skill supports user-defined overrides. See `/droid` skill § Skill Overrides.
 
+## Custom Instructions
+
+Any command accepts a ` -- {instruction}` suffix. Split on the **first** ` -- ` (space-dash-dash-space): left is the command and its args, right is additional context or instruction to carry through the entire execution.
+
+Example: `/project search droid -- show me what changed last week`
+
 If `projects_dir` is not configured, inform the user they need to set it up:
 ```bash
 mkdir -p ~/.droid/skills/project

package/dist/tools/project/skills/project/references/loading.md

@@ -32,6 +32,7 @@
    - Confirm which project was loaded
    - Summarize key context (2-3 sentences)
    - Use project contents for all subsequent work in the session
+   - **Do NOT read CHANGELOG.md** — it exists for on-demand lookup only. PROJECT.md already references it; that is enough context.
 
 7. **If `-- {instruction}` provided:** Execute the follow-up instruction against the loaded project
 

package/dist/tools/release/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "droid-release",
-  "version": "0.2.0",
+  "version": "0.2.2",
   "description": "Release ceremony automation — create release PRs, check status, notify Slack.",
   "author": {
     "name": "Orderful",

package/dist/tools/release/TOOL.yaml

@@ -1,6 +1,6 @@
 name: release
 description: "Release ceremony automation — create release PRs, check status, notify Slack."
-version: 0.2.0
+version: 0.2.2
 status: alpha
 
 includes:

package/dist/tools/release/skills/release/SKILL.md

@@ -45,6 +45,12 @@ droid config --get repos
 If no repos have `release_branch` set, tell user:
 > "No release repos configured. Run `droid repos add` and set a release branch to get started."
 
+## Custom Instructions
+
+Any command accepts a ` -- {instruction}` suffix. Split on the **first** ` -- ` (space-dash-dash-space): left is the command and its args, right is additional context or instruction to carry through the entire execution. Note: flag-style `--no-lock` uses `--flag` syntax (no surrounding spaces) and is distinct from this separator.
+
+Example: `/release start -- notify #releases-eng channel instead of the default`
+
 ## Commands
 
 | Command | Action |
@@ -78,3 +84,4 @@ See `references/templates.md` for Slack message and PR body templates.
 | Branch already locked | Warn user, offer to unlock first |
 | Release PR already exists | Show existing PR, ask if user wants to proceed |
 | CI checks failing | Show status, do not auto-merge |
+| High-risk label present at merge | Show extra confirmation prompt before standard merge confirmation; do not auto-skip |

package/dist/tools/release/skills/release/references/templates.md

@@ -23,6 +23,13 @@ Posted with :droid:
 
 If `--no-lock` was used, omit the lock line.
 
+If `HIGH_RISK_PRS` is non-empty, append the following block after `{pr_summary}` (before `Posted with :droid:`):
+```
+⚠️ *High-risk PRs:*
+- <{pr_url}|#{number}> {title} (@{author})
+```
+(one line per high-risk PR)
+
 Risk emojis:
 - Low Risk: `:large_green_circle:`
 - High Risk: `:warning:`
@@ -103,6 +110,16 @@ Where `{pr_list}` is a bulleted list of merged PRs:
 - #{number} {title} (@{author})
 ```
 
+If `HIGH_RISK_PRS` is non-empty, insert the following section between `{pr_list}` and the `---` divider:
+```markdown
+### ⚠️ High-Risk PRs
+
+The following PRs in this release are labelled `high-risk`. Coordinate with the authors before deploying.
+
+- #{number} {title} (@{author})
+```
+(one line per high-risk PR)
+
 ---
 
 ## Terminal Fallback
@@ -120,6 +137,11 @@ Release open for review — {repo_name}
 
 If `--no-lock` was used, omit the Lock line.
 
+If `HIGH_RISK_PRS` is non-empty, append:
+```
+  ⚠️  High-risk PRs: #{number} {title} (@{author}), ...
+```
+
 ### Release Merged (terminal)
 ```
 Release merged — {repo_name}