Task Analyzer Skill
Expert in analyzing natural language task descriptions and routing to appropriate craft commands.
When to Use
This skill is automatically activated when:
- User invokes
/craft:do <task>
- User describes a development task
- Context suggests multi-step workflow needed
Capabilities
Intent Recognition (D5, 2026-07-04: canonical source — /craft:do references
this table instead of duplicating it; see commands/do.md's Task Categories section)
Identifies task intent from natural language. Category names match /craft:do's
routing categories exactly (renamed 2026-07-04 to eliminate a naming drift where this
table used different labels — Create/Fix/Document/Refactor/Review — for the same 7
concepts /craft:do already routes on as Feature/Bug/Docs/Architecture, plus a
missing Quality category /craft:do has and this table didn't):
| Intent |
Keywords |
Example |
| Feature |
add, create, implement, build, new |
"add user authentication" |
| Bug |
fix, debug, resolve, repair, issue, error |
"fix the login bug" |
| Quality |
lint, quality, clean, improve |
"clean up the utils module" |
| Test |
test, verify, check, validate, coverage |
"test the api endpoints" |
| Docs |
document, docs, readme, changelog, explain |
"document the api" |
| Release |
release, deploy, publish, ship |
"prepare for release" |
| Architecture |
design, refactor, restructure, review, audit, analyze |
"refactor the utils module" |
Judgment call made here: the former standalone "Review" intent (review, audit,
analyze) had no corresponding category or command mapping in /craft:do at all — folded
its keywords into Architecture rather than inventing an 8th category /craft:do
would have no routing for.
Domain Detection
Identifies affected domain from context:
| Domain |
Indicators |
Commands |
| API |
api, endpoint, rest, graphql |
arch:plan, code:test-gen |
| UI |
ui, component, page, view |
frontend-designer skill |
| Database |
db, database, model, schema |
backend-designer skill |
| Auth |
auth, login, permission |
arch:plan, code:test-gen |
| Testing |
test, coverage, spec |
test, test --coverage |
| Docs |
docs, readme, changelog |
docs:sync, docs:validate |
Workflow Selection
Maps intent + domain to optimal command sequence:
Feature Workflow:
1. /craft:arch:plan - Design the feature
2. /craft:code:test-gen - Generate tests (TDD)
3. dev/git skill - Create feature branch (ask naturally; folded from /craft:git:branch)
4. → Implementation - User implements
5. /craft:test - Verify tests pass
6. dev/git skill - Commit/sync changes (ask naturally; folded from /craft:git:sync)
Bug Fix Workflow:
1. /craft:code:debug - Analyze the issue
2. /craft:test debug - Isolate in tests
3. → Fix - User implements fix
4. /craft:test - Verify fix
5. dev/git skill - Commit/sync changes (ask naturally; folded from /craft:git:sync)
Release Workflow:
1. /craft:code:deps-audit - Security scan
2. /craft:test release - Full test suite
3. /craft:code:lint release - Comprehensive lint
4. /craft:docs:changelog - Update changelog
5. /craft:code:release - Release workflow
Coded-Workflow Shape Detection (D7)
When a task description reads like a fixed coded shape —
decompose → cover N → verify M → synthesize — suggest
/craft:orch:workflow, do not silently switch to it.
Detect the shape conservatively (the helper
workflow_parse.detects_workflow_shape(text) fires only when ≥3 of the four
stage categories appear, or the explicit decompose…synthesize chain is
present):
| Stage category |
Trigger words |
| decompose |
decompose, break down, split into, dimensions, for each |
| fan-out |
fan out, in parallel, one per, cover each, per finding, reviewers |
| verify |
verify, verifier, double-check, confirm each |
| synthesize |
synthesize, aggregate, summarize, combine, merge findings |
Suggest, never switch (routing-false-positive guard). A lone "verify" or
"in parallel" must NOT route anywhere — auto-hijacking a task better served by
improvised orchestrate is the accepted residual risk. On a match, present a
suggestion the user confirms:
This looks like a fixed decompose → cover → verify → synthesize shape.
Consider: /craft:orch:workflow (coded, schema-gated, resumable)
Or keep going with improvised orchestration. Which do you want?
| Input |
Detected? |
Action |
| "decompose into dimensions, review each in parallel, verify, synthesize" |
yes |
suggest :workflow |
| "fan out reviewers, verify findings, then summarize" |
yes |
suggest :workflow |
| "verify the login flow works" |
no |
route normally |
| "review the architecture" |
no |
route normally |
Complexity Assessment
Determines task complexity for mode selection:
| Complexity |
Indicators |
Default Mode |
Max Parallel Sessions |
| Simple |
Single file, quick fix |
default |
5 |
| Medium |
Few files, feature |
default |
5 |
| Complex |
Many files, architecture |
debug |
no cap |
| Critical |
Release, security |
release |
no cap |
Advisory only — task-analyzer is used for command routing, not dispatch-time
enforcement. Nothing in orchestrator-v2/craft:orch:workflow reads this table
or blocks a dispatch that exceeds it (see #327).
Output Format
Analysis Report
╭─ Task Analysis ─────────────────────────────────────╮
│ Input: "add user authentication" │
├─────────────────────────────────────────────────────┤
│ Intent: Feature │
│ Domain: Authentication │
│ Complexity: Medium │
│ Workflow: Feature Development │
├─────────────────────────────────────────────────────┤
│ Recommended Commands: │
│ 1. /craft:arch:plan Design auth system │
│ 2. /craft:code:test-gen Generate auth tests │
│ 3. dev/git skill Create feature/auth │
╰─────────────────────────────────────────────────────╯
Integration
Works with:
/craft:do - Primary entry point/craft:check - Validation after workflows- All category commands (code, test, arch, etc.)
- All design skills (backend, frontend, devops)
Example Analyses
Simple Task
Input: "fix typo in readme"
Analysis:
- Intent: Bug
- Domain: Documentation
- Complexity: Simple
- Command: Direct edit (no routing needed)
Complex Task
Input: "implement oauth2 login with google"
Analysis:
- Intent: Feature
- Domain: Authentication + API
- Complexity: Complex
- Workflow: Feature Development
- Skills: backend-designer, devops-helper
- Commands: arch:plan → code:test-gen → git:branch
1---2name: task-analyzer3description: This skill should be used when the user asks to "analyze a task", "route a command", "what craft command should I use", or describes a development task that needs routing to the appropriate craft workflow. Analyzes natural language task descriptions and maps them to optimal craft command sequences.4---56# Task Analyzer Skill78Expert in analyzing natural language task descriptions and routing to appropriate craft commands.910## When to Use1112This skill is automatically activated when:1314- User invokes `/craft:do <task>`15- User describes a development task16- Context suggests multi-step workflow needed1718## Capabilities1920### Intent Recognition (D5, 2026-07-04: canonical source — `/craft:do` references2122this table instead of duplicating it; see `commands/do.md`'s Task Categories section)2324Identifies task intent from natural language. Category names match `/craft:do`'s25routing categories exactly (renamed 2026-07-04 to eliminate a naming drift where this26table used different labels — Create/Fix/Document/Refactor/Review — for the same 727concepts `/craft:do` already routes on as Feature/Bug/Docs/Architecture, plus a28missing **Quality** category `/craft:do` has and this table didn't):2930| Intent | Keywords | Example |31|--------|----------|---------|32| **Feature** | add, create, implement, build, new | "add user authentication" |33| **Bug** | fix, debug, resolve, repair, issue, error | "fix the login bug" |34| **Quality** | lint, quality, clean, improve | "clean up the utils module" |35| **Test** | test, verify, check, validate, coverage | "test the api endpoints" |36| **Docs** | document, docs, readme, changelog, explain | "document the api" |37| **Release** | release, deploy, publish, ship | "prepare for release" |38| **Architecture** | design, refactor, restructure, review, audit, analyze | "refactor the utils module" |3940**Judgment call made here:** the former standalone "Review" intent (review, audit,41analyze) had no corresponding category or command mapping in `/craft:do` at all — folded42its keywords into **Architecture** rather than inventing an 8th category `/craft:do`43would have no routing for.4445### Domain Detection4647Identifies affected domain from context:4849| Domain | Indicators | Commands |50|--------|------------|----------|51| **API** | api, endpoint, rest, graphql | arch:plan, code:test-gen |52| **UI** | ui, component, page, view | frontend-designer skill |53| **Database** | db, database, model, schema | backend-designer skill |54| **Auth** | auth, login, permission | arch:plan, code:test-gen |55| **Testing** | test, coverage, spec | test, test --coverage |56| **Docs** | docs, readme, changelog | docs:sync, docs:validate |5758### Workflow Selection5960Maps intent + domain to optimal command sequence:6162```63Feature Workflow:64 1. /craft:arch:plan - Design the feature65 2. /craft:code:test-gen - Generate tests (TDD)66 3. dev/git skill - Create feature branch (ask naturally; folded from /craft:git:branch)67 4. → Implementation - User implements68 5. /craft:test - Verify tests pass69 6. dev/git skill - Commit/sync changes (ask naturally; folded from /craft:git:sync)7071Bug Fix Workflow:72 1. /craft:code:debug - Analyze the issue73 2. /craft:test debug - Isolate in tests74 3. → Fix - User implements fix75 4. /craft:test - Verify fix76 5. dev/git skill - Commit/sync changes (ask naturally; folded from /craft:git:sync)7778Release Workflow:79 1. /craft:code:deps-audit - Security scan80 2. /craft:test release - Full test suite81 3. /craft:code:lint release - Comprehensive lint82 4. /craft:docs:changelog - Update changelog83 5. /craft:code:release - Release workflow84```8586### Coded-Workflow Shape Detection (D7)8788When a task description reads like a **fixed coded shape** —89decompose → cover N → verify M → synthesize — **suggest**90`/craft:orch:workflow`, do **not** silently switch to it.9192Detect the shape conservatively (the helper93`workflow_parse.detects_workflow_shape(text)` fires only when ≥3 of the four94stage categories appear, or the explicit `decompose…synthesize` chain is95present):9697| Stage category | Trigger words |98|----------------|---------------|99| decompose | decompose, break down, split into, dimensions, for each |100| fan-out | fan out, in parallel, one per, cover each, per finding, reviewers |101| verify | verify, verifier, double-check, confirm each |102| synthesize | synthesize, aggregate, summarize, combine, merge findings |103104**Suggest, never switch (routing-false-positive guard).** A lone "verify" or105"in parallel" must NOT route anywhere — auto-hijacking a task better served by106improvised `orchestrate` is the accepted residual risk. On a match, present a107*suggestion* the user confirms:108109```110This looks like a fixed decompose → cover → verify → synthesize shape.111Consider: /craft:orch:workflow (coded, schema-gated, resumable)112Or keep going with improvised orchestration. Which do you want?113```114115| Input | Detected? | Action |116|-------|-----------|--------|117| "decompose into dimensions, review each in parallel, verify, synthesize" | yes | suggest `:workflow` |118| "fan out reviewers, verify findings, then summarize" | yes | suggest `:workflow` |119| "verify the login flow works" | no | route normally |120| "review the architecture" | no | route normally |121122### Complexity Assessment123124Determines task complexity for mode selection:125126| Complexity | Indicators | Default Mode | Max Parallel Sessions |127|------------|------------|--------------|------------------------|128| **Simple** | Single file, quick fix | default | 5 |129| **Medium** | Few files, feature | default | 5 |130| **Complex** | Many files, architecture | debug | no cap |131| **Critical** | Release, security | release | no cap |132133> Advisory only — `task-analyzer` is used for command routing, not dispatch-time134> enforcement. Nothing in `orchestrator-v2`/`craft:orch:workflow` reads this table135> or blocks a dispatch that exceeds it (see #327).136137## Output Format138139### Analysis Report140141```142╭─ Task Analysis ─────────────────────────────────────╮143│ Input: "add user authentication" │144├─────────────────────────────────────────────────────┤145│ Intent: Feature │146│ Domain: Authentication │147│ Complexity: Medium │148│ Workflow: Feature Development │149├─────────────────────────────────────────────────────┤150│ Recommended Commands: │151│ 1. /craft:arch:plan Design auth system │152│ 2. /craft:code:test-gen Generate auth tests │153│ 3. dev/git skill Create feature/auth │154╰─────────────────────────────────────────────────────╯155```156157## Integration158159Works with:160161- `/craft:do` - Primary entry point162- `/craft:check` - Validation after workflows163- All category commands (code, test, arch, etc.)164- All design skills (backend, frontend, devops)165166## Example Analyses167168### Simple Task169170```171Input: "fix typo in readme"172Analysis:173 - Intent: Bug174 - Domain: Documentation175 - Complexity: Simple176 - Command: Direct edit (no routing needed)177```178179### Complex Task180181```182Input: "implement oauth2 login with google"183Analysis:184 - Intent: Feature185 - Domain: Authentication + API186 - Complexity: Complex187 - Workflow: Feature Development188 - Skills: backend-designer, devops-helper189 - Commands: arch:plan → code:test-gen → git:branch190```
Data-Wise