@dv.nghiem/flowdeck 0.3.8 → 0.4.0
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/README.md +13 -122
- package/dist/agents/code-explorer.d.ts.map +1 -1
- package/dist/agents/mapper.d.ts.map +1 -1
- package/dist/agents/orchestrator.d.ts.map +1 -1
- package/dist/agents/planner.d.ts.map +1 -1
- package/dist/agents/specialist.d.ts.map +1 -1
- package/dist/dashboard/server.mjs +12 -2
- package/dist/hooks/compaction-hook.d.ts.map +1 -1
- package/dist/hooks/file-tracker.d.ts +6 -0
- package/dist/hooks/file-tracker.d.ts.map +1 -1
- package/dist/hooks/notifications.d.ts.map +1 -1
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +752 -785
- package/dist/lib/completion-validator.d.ts +51 -0
- package/dist/lib/completion-validator.d.ts.map +1 -0
- package/dist/lib/recommended-question.d.ts +24 -0
- package/dist/lib/recommended-question.d.ts.map +1 -0
- package/dist/lib/research-gate.d.ts +97 -0
- package/dist/lib/research-gate.d.ts.map +1 -0
- package/dist/lib/research-gate.test.d.ts +2 -0
- package/dist/lib/research-gate.test.d.ts.map +1 -0
- package/dist/mcp/index.d.ts +14 -3
- package/dist/mcp/index.d.ts.map +1 -1
- package/dist/services/codegraph.d.ts +36 -0
- package/dist/services/codegraph.d.ts.map +1 -0
- package/dist/services/codegraph.test.d.ts +2 -0
- package/dist/services/codegraph.test.d.ts.map +1 -0
- package/dist/services/command-validator.d.ts +11 -0
- package/dist/services/command-validator.d.ts.map +1 -1
- package/dist/services/preflight-explorer.d.ts +130 -0
- package/dist/services/preflight-explorer.d.ts.map +1 -0
- package/dist/services/preflight-explorer.test.d.ts +25 -0
- package/dist/services/preflight-explorer.test.d.ts.map +1 -0
- package/dist/services/question-guard.d.ts +96 -0
- package/dist/services/question-guard.d.ts.map +1 -0
- package/dist/services/quick-router.d.ts +40 -1
- package/dist/services/quick-router.d.ts.map +1 -1
- package/dist/services/recommended-question.test.d.ts +2 -0
- package/dist/services/recommended-question.test.d.ts.map +1 -0
- package/dist/services/supervisor-binding.d.ts +3 -1
- package/dist/services/supervisor-binding.d.ts.map +1 -1
- package/dist/tools/codebase-index.d.ts +30 -0
- package/dist/tools/codebase-index.d.ts.map +1 -0
- package/dist/tools/codebase-index.test.d.ts +2 -0
- package/dist/tools/codebase-index.test.d.ts.map +1 -0
- package/dist/tools/codegraph-tool.d.ts +3 -0
- package/dist/tools/codegraph-tool.d.ts.map +1 -0
- package/dist/tools/planning-state-lib.d.ts +23 -0
- package/dist/tools/planning-state-lib.d.ts.map +1 -1
- package/docs/agents/index.md +154 -0
- package/docs/commands/fd-ask.md +71 -39
- package/docs/commands/fd-checkpoint.md +63 -8
- package/docs/commands/fd-deploy-check.md +166 -9
- package/docs/commands/fd-design.md +101 -0
- package/docs/commands/fd-discuss.md +87 -20
- package/docs/commands/fd-doctor.md +100 -13
- package/docs/commands/fd-done.md +215 -0
- package/docs/commands/fd-execute.md +104 -0
- package/docs/commands/fd-fix-bug.md +144 -24
- package/docs/commands/fd-map-codebase.md +85 -21
- package/docs/commands/fd-multi-repo.md +155 -40
- package/docs/commands/fd-new-feature.md +63 -19
- package/docs/commands/fd-plan.md +80 -27
- package/docs/commands/fd-quick.md +143 -29
- package/docs/commands/fd-reflect.md +81 -13
- package/docs/commands/fd-resume.md +65 -8
- package/docs/commands/fd-status.md +80 -12
- package/docs/commands/fd-suggest.md +114 -0
- package/docs/commands/fd-translate-intent.md +69 -9
- package/docs/commands/fd-verify.md +71 -14
- package/docs/commands/fd-write-docs.md +121 -8
- package/docs/concepts/architecture.md +163 -0
- package/docs/concepts/governance.md +242 -0
- package/docs/concepts/intelligence.md +145 -0
- package/docs/concepts/multi-repo.md +227 -0
- package/docs/concepts/workflows.md +205 -0
- package/docs/configuration/index.md +208 -0
- package/docs/configuration/opencode-settings.md +98 -0
- package/docs/getting-started/first-project.md +126 -0
- package/docs/getting-started/installation.md +73 -0
- package/docs/getting-started/quick-start.md +74 -0
- package/docs/index.md +36 -72
- package/docs/reference/hooks.md +176 -0
- package/docs/reference/rules.md +109 -0
- package/docs/skills/code-review.md +47 -0
- package/docs/skills/index.md +148 -0
- package/docs/skills/planning.md +39 -0
- package/package.json +1 -1
- package/src/commands/fd-deploy-check.md +2 -2
- package/src/commands/fd-discuss.md +128 -7
- package/src/commands/fd-done.md +196 -0
- package/src/commands/fd-execute.md +43 -6
- package/src/commands/fd-fix-bug.md +43 -6
- package/src/commands/fd-map-codebase.md +100 -20
- package/src/commands/fd-multi-repo.md +1 -1
- package/src/commands/fd-new-feature.md +14 -5
- package/src/commands/fd-plan.md +38 -1
- package/src/commands/fd-quick.md +77 -14
- package/src/commands/fd-resume.md +1 -1
- package/src/commands/fd-status.md +1 -1
- package/src/commands/fd-verify.md +16 -2
- package/src/commands/fd-write-docs.md +30 -5
- package/src/rules/common/behavioral.md +63 -0
- package/src/skills/codebase-mapping/SKILL.md +1 -1
- package/src/skills/context-load/SKILL.md +1 -1
- package/src/skills/multi-repo/SKILL.md +1 -1
- package/src/skills/repo-memory-graph/SKILL.md +1 -1
- package/dist/hooks/memory-hook.d.ts +0 -28
- package/dist/hooks/memory-hook.d.ts.map +0 -1
- package/dist/services/memory-store.d.ts +0 -73
- package/dist/services/memory-store.d.ts.map +0 -1
- package/dist/services/memory-store.test.d.ts +0 -2
- package/dist/services/memory-store.test.d.ts.map +0 -1
- package/dist/tools/memory-search.d.ts +0 -3
- package/dist/tools/memory-search.d.ts.map +0 -1
- package/dist/tools/memory-status.d.ts +0 -3
- package/dist/tools/memory-status.d.ts.map +0 -1
- package/docs/USER_GUIDE.md +0 -20
- package/docs/agents.md +0 -544
- package/docs/best-practices.md +0 -47
- package/docs/commands/fd-new-project.md +0 -24
- package/docs/commands.md +0 -557
- package/docs/configuration.md +0 -325
- package/docs/design-first-workflow.md +0 -94
- package/docs/feature-integration-architecture.md +0 -227
- package/docs/installation.md +0 -123
- package/docs/intelligence.md +0 -370
- package/docs/memory.md +0 -69
- package/docs/multi-repo.md +0 -201
- package/docs/notifications.md +0 -170
- package/docs/optimization-baseline.md +0 -21
- package/docs/quick-start.md +0 -197
- package/docs/rules.md +0 -432
- package/docs/skills.md +0 -417
- package/docs/workflows.md +0 -134
- package/src/commands/fd-new-project.md +0 -114
|
@@ -1,23 +1,91 @@
|
|
|
1
|
-
|
|
2
|
-
description: Post-session reflection and skill capture — analyse artifacts, propose improvements, capture session patterns
|
|
3
|
-
argument-hint: [--mode=reflect,learn]
|
|
4
|
-
---
|
|
1
|
+
# /fd-reflect
|
|
5
2
|
|
|
6
|
-
|
|
3
|
+
**Purpose:** Post-session reflection and skill capture — analyze session artifacts, propose improvements to FlowDeck's knowledge base, and optionally capture reusable skills from patterns discovered during the session.
|
|
7
4
|
|
|
8
|
-
|
|
5
|
+
## Usage
|
|
9
6
|
|
|
10
|
-
|
|
7
|
+
/fd-reflect [--mode=reflect,learn]
|
|
11
8
|
|
|
12
|
-
|
|
13
|
-
Analyzes session artifacts, proposes skills/policies/workflow changes.
|
|
9
|
+
## What Happens
|
|
14
10
|
|
|
15
|
-
###
|
|
16
|
-
|
|
11
|
+
### Reflect Mode (default)
|
|
12
|
+
|
|
13
|
+
1. **Gather artifacts.** Call the `reflect` tool to collect session artifacts and generate reflection context.
|
|
14
|
+
|
|
15
|
+
2. **Analyze the session.**
|
|
16
|
+
- Which tools were called most often? Any redundant calls?
|
|
17
|
+
- Which decisions were made? Do they reveal a repeatable pattern?
|
|
18
|
+
- Were there failures? What caused them?
|
|
19
|
+
- What knowledge was absent and had to be worked out from scratch?
|
|
20
|
+
|
|
21
|
+
3. **Produce improvement proposals.**
|
|
22
|
+
- **New skill** → call `create-skill` to capture it in `src/skills/`
|
|
23
|
+
- **Policy** → propose a new entry in `.codebase/POLICIES.json` for user review
|
|
24
|
+
- **Workflow change** → note it clearly for user to decide
|
|
25
|
+
|
|
26
|
+
4. **Execute skill creation** for any new skills identified.
|
|
27
|
+
|
|
28
|
+
5. **Final report.** Provide:
|
|
29
|
+
- What was captured (new skills created)
|
|
30
|
+
- What requires human review (policy proposals, workflow changes)
|
|
31
|
+
- 3-5 bullet summary of the session's most impactful learnings
|
|
32
|
+
|
|
33
|
+
### Learn Mode (`--mode=learn`)
|
|
34
|
+
|
|
35
|
+
1. **Identify worth capturing.** Look for:
|
|
36
|
+
- Novel problems requiring non-obvious solutions
|
|
37
|
+
- Patterns that needed human guidance to resolve
|
|
38
|
+
- Workflows or sequences that would save time if remembered
|
|
39
|
+
- Pitfalls that were hit and corrected
|
|
40
|
+
|
|
41
|
+
If nothing significant was found, reply: "No new patterns to capture from this session." and stop.
|
|
42
|
+
|
|
43
|
+
2. **Draft the skill.** Structure as:
|
|
44
|
+
- `## When to Activate` — concrete triggers (e.g., "when X file pattern exists", "when user asks about Y")
|
|
45
|
+
- `## Steps` — ordered, concrete steps to apply the skill
|
|
46
|
+
- `## Examples` — at least one short concrete example
|
|
47
|
+
- `## Pitfalls` — common mistakes to avoid
|
|
48
|
+
|
|
49
|
+
3. **Choose a name.** Use `$ARGUMENTS` if provided as skill name; otherwise derive a kebab-case name from the pattern.
|
|
50
|
+
|
|
51
|
+
4. **Write the skill** using the `create-skill` tool with:
|
|
52
|
+
- `name`: kebab-case identifier
|
|
53
|
+
- `description`: one sentence summary
|
|
54
|
+
- `content`: full Markdown body
|
|
55
|
+
- `tags`: 2-4 relevant tags
|
|
56
|
+
|
|
57
|
+
5. **Report** what was captured, why it is useful, and remind the user to restart OpenCode to activate it.
|
|
58
|
+
|
|
59
|
+
## Output / State
|
|
60
|
+
|
|
61
|
+
Files created (Learn mode):
|
|
62
|
+
- `src/skills/<name>.md` — new skill file
|
|
63
|
+
|
|
64
|
+
Files modified (Reflect mode):
|
|
65
|
+
- `.codebase/POLICIES.json` — proposed policy entries (user review needed)
|
|
17
66
|
|
|
18
67
|
## Examples
|
|
19
68
|
|
|
20
69
|
```
|
|
21
70
|
/fd-reflect
|
|
22
|
-
|
|
23
|
-
|
|
71
|
+
```
|
|
72
|
+
|
|
73
|
+
Run post-session reflection and produce improvement proposals.
|
|
74
|
+
|
|
75
|
+
```
|
|
76
|
+
/fd-reflect --mode=learn
|
|
77
|
+
```
|
|
78
|
+
|
|
79
|
+
Capture the most significant pattern from this session as a reusable skill.
|
|
80
|
+
|
|
81
|
+
```
|
|
82
|
+
/fd-reflect --mode=learn api-error-handling
|
|
83
|
+
```
|
|
84
|
+
|
|
85
|
+
Capture the session pattern as a skill named "api-error-handling".
|
|
86
|
+
|
|
87
|
+
## Related Commands
|
|
88
|
+
|
|
89
|
+
- `/fd-map-codebase` — map the codebase before starting a feature
|
|
90
|
+
- `/fd-execute` — run implementation (reflection happens after)
|
|
91
|
+
- `/fd-checkpoint` — save state before ending a session
|
|
@@ -1,10 +1,67 @@
|
|
|
1
|
-
|
|
2
|
-
description: Resume from last checkpoint — reloads STATE.md and PLAN.md context
|
|
3
|
-
---
|
|
4
|
-
Run the FlowDeck resume workflow to reload project context and continue from where you stopped.
|
|
1
|
+
# /fd-resume
|
|
5
2
|
|
|
6
|
-
|
|
3
|
+
**Purpose:** Reload STATE.md, PLAN.md, and DISCUSS.md to continue an interrupted session — brief the user, PAUSE for confirmation, then resume from where work stopped.
|
|
7
4
|
|
|
8
|
-
|
|
9
|
-
|
|
10
|
-
|
|
5
|
+
## Usage
|
|
6
|
+
|
|
7
|
+
/fd-resume [--yes]
|
|
8
|
+
|
|
9
|
+
## What Happens
|
|
10
|
+
|
|
11
|
+
1. **Pre-flight check.**
|
|
12
|
+
- Verify `.planning/STATE.md` exists — error if not ("No active feature. Run `/fd-map-codebase` then `/fd-new-feature` to start a feature.")
|
|
13
|
+
|
|
14
|
+
2. **Read and parse STATE.md.** Extract phase, status, last_updated, plan_confirmed.
|
|
15
|
+
|
|
16
|
+
3. **Read PLAN.md** (`.planning/phases/phase-<N>/PLAN.md`) if it exists — show preview (first 20 lines).
|
|
17
|
+
|
|
18
|
+
4. **Read DISCUSS.md** (`.planning/phases/phase-<N>/DISCUSS.md`) if it exists — show decision count.
|
|
19
|
+
|
|
20
|
+
5. **Present session summary:**
|
|
21
|
+
|
|
22
|
+
```
|
|
23
|
+
═══════════════════════════════════════════════
|
|
24
|
+
RESUMING SESSION
|
|
25
|
+
═══════════════════════════════════════════════
|
|
26
|
+
Phase: <N> | Status: <status>
|
|
27
|
+
Last updated: <timestamp>
|
|
28
|
+
Plan confirmed: <yes/no>
|
|
29
|
+
Decisions: <X> from DISCUSS.md
|
|
30
|
+
|
|
31
|
+
Plan preview:
|
|
32
|
+
<first 10 lines of PLAN.md>
|
|
33
|
+
───────────────────────────────────────────────
|
|
34
|
+
Type CONFIRM to resume execution from this point.
|
|
35
|
+
═══════════════════════════════════════════════
|
|
36
|
+
```
|
|
37
|
+
|
|
38
|
+
6. **PAUSE for confirmation** (unless `--yes` is passed). Wait for user to type CONFIRM.
|
|
39
|
+
|
|
40
|
+
7. **After confirmation:**
|
|
41
|
+
- If `plan_confirmed: true` and PLAN.md has uncompleted steps → proceed with implementation
|
|
42
|
+
- If no plan exists → suggest running `/fd-plan`
|
|
43
|
+
- Brief the user on what the next step is before starting
|
|
44
|
+
|
|
45
|
+
## Output / State
|
|
46
|
+
|
|
47
|
+
No new files created. Resumes from existing STATE.md and PLAN.md.
|
|
48
|
+
|
|
49
|
+
## Examples
|
|
50
|
+
|
|
51
|
+
```
|
|
52
|
+
/fd-resume
|
|
53
|
+
```
|
|
54
|
+
|
|
55
|
+
Show session summary and wait for CONFIRM before resuming.
|
|
56
|
+
|
|
57
|
+
```
|
|
58
|
+
/fd-resume --yes
|
|
59
|
+
```
|
|
60
|
+
|
|
61
|
+
Skip confirmation and immediately resume from the last checkpoint.
|
|
62
|
+
|
|
63
|
+
## Related Commands
|
|
64
|
+
|
|
65
|
+
- `/fd-checkpoint` — save a checkpoint before closing a session
|
|
66
|
+
- `/fd-plan` — create a plan if no PLAN.md exists to resume
|
|
67
|
+
- `/fd-execute` — continue implementation (auto-triggered after CONFIRM)
|
|
@@ -1,31 +1,99 @@
|
|
|
1
|
-
|
|
2
|
-
description: View project status — combined status, roadmap, workspace overview, and progress
|
|
3
|
-
argument-hint: [--roadmap | --workspace | --phase=N]
|
|
4
|
-
---
|
|
1
|
+
# /fd-status
|
|
5
2
|
|
|
6
|
-
|
|
3
|
+
**Purpose:** View project progress, roadmap phase statuses, and workspace overview — combined status display with optional detailed flags.
|
|
7
4
|
|
|
8
|
-
|
|
5
|
+
## Usage
|
|
9
6
|
|
|
10
|
-
|
|
7
|
+
/fd-status [--roadmap | --workspace | --phase=N]
|
|
8
|
+
|
|
9
|
+
## What Happens
|
|
11
10
|
|
|
12
11
|
### Default (no flags)
|
|
13
|
-
|
|
12
|
+
|
|
13
|
+
Reads `.planning/STATE.md` and displays:
|
|
14
|
+
```
|
|
15
|
+
Phase: <N> | Status: <status> | Updated: <timestamp>
|
|
16
|
+
────────────────────────────────────────────────────────────
|
|
17
|
+
Plan: <X> steps (<Y> complete)
|
|
18
|
+
Plan confirmed: <yes/no>
|
|
19
|
+
```
|
|
14
20
|
|
|
15
21
|
### Roadmap (`--roadmap`)
|
|
16
|
-
|
|
22
|
+
|
|
23
|
+
Reads `.planning/ROADMAP.md` and `.planning/STATE.md` and displays:
|
|
24
|
+
```
|
|
25
|
+
═══════════════════════════════════════
|
|
26
|
+
PROJECT ROADMAP
|
|
27
|
+
═══════════════════════════════════════
|
|
28
|
+
✅ Phase 1: <name> — completed
|
|
29
|
+
🔄 Phase 2: <name> — in progress ← current
|
|
30
|
+
⏳ Phase 3: <name> — planned
|
|
31
|
+
═══════════════════════════════════════
|
|
32
|
+
```
|
|
17
33
|
|
|
18
34
|
### Workspace (`--workspace`)
|
|
19
|
-
|
|
35
|
+
|
|
36
|
+
Reads `.planning/config.json` for registered repositories and each repo's STATE.md, displaying:
|
|
37
|
+
```
|
|
38
|
+
════════════════════════════════════════════════════
|
|
39
|
+
WORKSPACE OVERVIEW
|
|
40
|
+
════════════════════════════════════════════════════
|
|
41
|
+
frontend — Phase 2 | in_progress | Plan: ✅ | Updated: <time>
|
|
42
|
+
backend — Phase 3 | completed | Plan: ✅ | Updated: <time>
|
|
43
|
+
shared — Phase 1 | planned | Plan: ❌ | Updated: <time>
|
|
44
|
+
────────────────────────────────────────────────────
|
|
45
|
+
Total: 3 repos | 1 in progress | 1 completed | 1 planned
|
|
46
|
+
════════════════════════════════════════════════════
|
|
47
|
+
```
|
|
20
48
|
|
|
21
49
|
### Phase Detail (`--phase=N`)
|
|
22
|
-
|
|
50
|
+
|
|
51
|
+
Reads the specific phase's STATE.md and PLAN.md (if exists) and displays:
|
|
52
|
+
```
|
|
53
|
+
Phase <N> Detail
|
|
54
|
+
Status: <status>
|
|
55
|
+
Plan file: <path>
|
|
56
|
+
Plan confirmed: <yes/no>
|
|
57
|
+
|
|
58
|
+
Steps:
|
|
59
|
+
✅ Step 1: <name> — completed
|
|
60
|
+
🔄 Step 2: <name> — in progress
|
|
61
|
+
⬜ Step 3: <name> — pending
|
|
62
|
+
⬜ Step 4: <name> — pending
|
|
63
|
+
```
|
|
64
|
+
|
|
65
|
+
## Output / State
|
|
66
|
+
|
|
67
|
+
No files modified. Read-only display command.
|
|
23
68
|
|
|
24
69
|
## Examples
|
|
25
70
|
|
|
26
71
|
```
|
|
27
72
|
/fd-status
|
|
73
|
+
```
|
|
74
|
+
|
|
75
|
+
Show current phase summary.
|
|
76
|
+
|
|
77
|
+
```
|
|
28
78
|
/fd-status --roadmap
|
|
79
|
+
```
|
|
80
|
+
|
|
81
|
+
Show full project roadmap with all phase statuses.
|
|
82
|
+
|
|
83
|
+
```
|
|
29
84
|
/fd-status --workspace
|
|
85
|
+
```
|
|
86
|
+
|
|
87
|
+
Show overview of all registered repositories in the workspace.
|
|
88
|
+
|
|
89
|
+
```
|
|
30
90
|
/fd-status --phase=2
|
|
31
|
-
```
|
|
91
|
+
```
|
|
92
|
+
|
|
93
|
+
Show detailed progress for phase 2.
|
|
94
|
+
|
|
95
|
+
## Related Commands
|
|
96
|
+
|
|
97
|
+
- `/fd-map-codebase` — map the codebase (required before starting a feature)
|
|
98
|
+
- `/fd-execute` — advance phase status by implementing steps
|
|
99
|
+
- `/fd-verify` — update phase status to `verified` after full pipeline pass
|
|
@@ -0,0 +1,114 @@
|
|
|
1
|
+
# /fd-suggest
|
|
2
|
+
|
|
3
|
+
**Purpose:** Analyze the codebase and suggest high-value feature opportunities with implementation guidance.
|
|
4
|
+
|
|
5
|
+
## Usage
|
|
6
|
+
|
|
7
|
+
/fd-suggest [--category=TYPE] [--limit=N]
|
|
8
|
+
|
|
9
|
+
## Arguments
|
|
10
|
+
|
|
11
|
+
- `--category=TYPE` (optional) — focus on specific type: `tools`, `hooks`, `agents`, or `skills`
|
|
12
|
+
- `--limit=N` (optional) — number of suggestions to return (default: 5)
|
|
13
|
+
|
|
14
|
+
## What Happens
|
|
15
|
+
|
|
16
|
+
### Step 1: Detect Environment
|
|
17
|
+
|
|
18
|
+
Check if `.planning/` directory exists:
|
|
19
|
+
- If it exists: Use current project context, analyze within existing architecture
|
|
20
|
+
- If not: This is a standalone analysis, scan the codebase root
|
|
21
|
+
|
|
22
|
+
### Step 2: Run Analysis Tools
|
|
23
|
+
|
|
24
|
+
Execute tools to gather insight data:
|
|
25
|
+
|
|
26
|
+
1. **Volatility Map** — `volatility-map` tool to find high-change areas (instability signals)
|
|
27
|
+
2. **Decision Trace** — `decision-trace` tool to find incomplete or TODO decisions
|
|
28
|
+
3. **Failure Replay** — `failure-replay` tool to find recurring failure patterns
|
|
29
|
+
4. **Codebase State** — `codebase-state` tool to get overall project structure
|
|
30
|
+
|
|
31
|
+
### Step 3: Scan for Gaps
|
|
32
|
+
|
|
33
|
+
Analyze the codebase structure to identify gaps:
|
|
34
|
+
|
|
35
|
+
- **Tools** — Are all tool categories covered? (planning, state, execution, analysis)
|
|
36
|
+
- **Hooks** — Are lifecycle hooks comprehensive?
|
|
37
|
+
- **Agents** — Are agent types sufficient for common tasks?
|
|
38
|
+
- **Skills** — Are there skill coverage gaps for major frameworks?
|
|
39
|
+
|
|
40
|
+
### Step 4: Generate Suggestions
|
|
41
|
+
|
|
42
|
+
For each valid suggestion (up to --limit), produce:
|
|
43
|
+
```
|
|
44
|
+
## Suggestion N: [Feature Name]
|
|
45
|
+
|
|
46
|
+
**Category:** tools|hooks|agents|skills
|
|
47
|
+
**Impact:** HIGH|MEDIUM|LOW
|
|
48
|
+
**Complexity:** HIGH|MEDIUM|LOW
|
|
49
|
+
|
|
50
|
+
### Problem Statement
|
|
51
|
+
[What issue this solves, with evidence from analysis]
|
|
52
|
+
|
|
53
|
+
### Proposed Solution
|
|
54
|
+
[Brief description of the approach]
|
|
55
|
+
|
|
56
|
+
### Implementation Steps
|
|
57
|
+
1. [Step with TDD approach - write test first]
|
|
58
|
+
2. [Step referencing existing patterns]
|
|
59
|
+
3. [Step with integration points]
|
|
60
|
+
|
|
61
|
+
### Files Affected
|
|
62
|
+
- `src/tools/xxx.ts` or `src/hooks/xxx.ts` or `src/agents/xxx.ts`
|
|
63
|
+
- `src/skills/xxx/SKILL.md` (if applicable)
|
|
64
|
+
- `docs/xxx.md` (if applicable)
|
|
65
|
+
|
|
66
|
+
### Integration
|
|
67
|
+
- How this connects to existing FlowDeck tools/agents
|
|
68
|
+
- Any new dependencies or configuration needed
|
|
69
|
+
```
|
|
70
|
+
|
|
71
|
+
### Step 5: Rank and Present
|
|
72
|
+
|
|
73
|
+
Sort suggestions by impact/complexity ratio (highest first).
|
|
74
|
+
|
|
75
|
+
## Output / State
|
|
76
|
+
|
|
77
|
+
```
|
|
78
|
+
# Feature Suggestions
|
|
79
|
+
|
|
80
|
+
Found N suggestions ranked by impact:
|
|
81
|
+
|
|
82
|
+
[Suggestions 1-N]
|
|
83
|
+
|
|
84
|
+
---
|
|
85
|
+
Run /fd-discuss [topic] to start a discussion on any suggestion.
|
|
86
|
+
```
|
|
87
|
+
|
|
88
|
+
## Examples
|
|
89
|
+
|
|
90
|
+
**Get 5 feature suggestions (default):**
|
|
91
|
+
```
|
|
92
|
+
/fd-suggest
|
|
93
|
+
```
|
|
94
|
+
|
|
95
|
+
**Get 10 suggestions:**
|
|
96
|
+
```
|
|
97
|
+
/fd-suggest --limit=10
|
|
98
|
+
```
|
|
99
|
+
|
|
100
|
+
**Focus on tool gaps only:**
|
|
101
|
+
```
|
|
102
|
+
/fd-suggest --category=tools
|
|
103
|
+
```
|
|
104
|
+
|
|
105
|
+
**Find agent-related opportunities:**
|
|
106
|
+
```
|
|
107
|
+
/fd-suggest --category=agents
|
|
108
|
+
```
|
|
109
|
+
|
|
110
|
+
## Related Commands
|
|
111
|
+
|
|
112
|
+
- `/fd-discuss` — explore any suggestion in depth
|
|
113
|
+
- `/fd-plan` — plan implementation of a selected suggestion
|
|
114
|
+
- `/fd-translate-intent` — convert a vague request into concrete options
|
|
@@ -1,17 +1,77 @@
|
|
|
1
|
-
|
|
2
|
-
description: Intent-to-Change Translator — convert vague requests into 3–5 concrete, ranked implementation options with tradeoffs
|
|
3
|
-
argument-hint: [vague intent, e.g. "make checkout faster"]
|
|
4
|
-
---
|
|
1
|
+
# /fd-translate-intent
|
|
5
2
|
|
|
6
|
-
|
|
3
|
+
**Purpose:** Convert vague requests into ranked concrete implementation options with tradeoffs.
|
|
7
4
|
|
|
8
|
-
|
|
5
|
+
## Usage
|
|
6
|
+
|
|
7
|
+
/fd-translate-intent [vague intent, e.g. "make checkout faster"]
|
|
8
|
+
|
|
9
|
+
## What Happens
|
|
10
|
+
|
|
11
|
+
Run two agents in parallel:
|
|
12
|
+
|
|
13
|
+
- **@architect**: Decomposes the input into 3-5 concrete implementation options. For each option provides:
|
|
14
|
+
- **Name**: short label
|
|
15
|
+
- **Description**: what exactly would be changed
|
|
16
|
+
- **Files affected**: list of files/modules that would change
|
|
17
|
+
- **Effort**: S (hours) / M (days) / L (week+)
|
|
18
|
+
- **Risk**: low / medium / high
|
|
19
|
+
- **Tradeoffs**: pros and cons
|
|
20
|
+
|
|
21
|
+
- **@researcher**: For each option, fetches relevant codebase context — finds existing patterns, prior art, and constraints that apply
|
|
22
|
+
|
|
23
|
+
## Output / State
|
|
24
|
+
|
|
25
|
+
Report format:
|
|
26
|
+
```
|
|
27
|
+
════════════════════════════════════════════
|
|
28
|
+
INTENT TRANSLATION: "$ARGUMENTS"
|
|
29
|
+
════════════════════════════════════════════
|
|
30
|
+
|
|
31
|
+
Option 1 (Recommended): <name>
|
|
32
|
+
Description: <what changes>
|
|
33
|
+
Files: <list>
|
|
34
|
+
Effort: <S|M|L> | Risk: <low|med|high>
|
|
35
|
+
Pros: <pros>
|
|
36
|
+
Cons: <cons>
|
|
37
|
+
|
|
38
|
+
Option 2: <name>
|
|
39
|
+
...
|
|
40
|
+
|
|
41
|
+
Option 3: <name>
|
|
42
|
+
...
|
|
43
|
+
|
|
44
|
+
────────────────────────────────────────────
|
|
45
|
+
Clarifying Questions:
|
|
46
|
+
1. <question about unclear aspect>
|
|
47
|
+
2. <question>
|
|
48
|
+
|
|
49
|
+
Assumptions Made:
|
|
50
|
+
- <assumption>
|
|
51
|
+
════════════════════════════════════════════
|
|
52
|
+
```
|
|
53
|
+
|
|
54
|
+
The command presents all options and asks: "Which option would you like to proceed with?"
|
|
9
55
|
|
|
10
56
|
## Examples
|
|
11
57
|
|
|
58
|
+
**Convert a vague request:**
|
|
59
|
+
```
|
|
60
|
+
/fd-translate-intent "make checkout faster"
|
|
12
61
|
```
|
|
13
|
-
|
|
14
|
-
|
|
62
|
+
|
|
63
|
+
**Translate a high-level request:**
|
|
15
64
|
```
|
|
65
|
+
/fd-translate-intent "improve auth security"
|
|
66
|
+
```
|
|
67
|
+
|
|
68
|
+
**Clarify a feature direction:**
|
|
69
|
+
```
|
|
70
|
+
/fd-translate-intent "add mobile support"
|
|
71
|
+
```
|
|
72
|
+
|
|
73
|
+
## Related Commands
|
|
16
74
|
|
|
17
|
-
|
|
75
|
+
- `/fd-discuss` — explore the problem space before translating intent
|
|
76
|
+
- `/fd-plan` — plan the selected implementation option
|
|
77
|
+
- `/fd-execute` — execute the chosen implementation
|
|
@@ -1,18 +1,75 @@
|
|
|
1
|
-
|
|
2
|
-
description: Review code quality, security, and conventions — runs parallel @reviewer + @security-auditor agents
|
|
3
|
-
argument-hint: "[file-pattern|dir|stage number]"
|
|
4
|
-
---
|
|
1
|
+
# /fd-verify
|
|
5
2
|
|
|
6
|
-
|
|
7
|
-
- **Specific files or directories** — `fd-verify src/services/`
|
|
8
|
-
- **Git staged changes** — `fd-verify staged`
|
|
9
|
-
- **Full project** — `fd-verify`
|
|
3
|
+
**Purpose:** Full verification pipeline after feature implementation — runs tests, code review, security scan, and deploy check; updates STATE.md to `verified` on full pass.
|
|
10
4
|
|
|
11
|
-
##
|
|
5
|
+
## Usage
|
|
12
6
|
|
|
13
|
-
|
|
14
|
-
2. **Create deployment checkpoint** → `/fd-checkpoint`
|
|
15
|
-
3. **Deploy to production** → `/fd-deploy-check`
|
|
16
|
-
4. **View project dashboard** → `/fd-dashboard`
|
|
7
|
+
/fd-verify [--phase=N] [--env=staging|production]
|
|
17
8
|
|
|
18
|
-
|
|
9
|
+
## What Happens
|
|
10
|
+
|
|
11
|
+
1. **Pre-flight checks.**
|
|
12
|
+
- Verify `.planning/` and STATE.md exist
|
|
13
|
+
- Read current phase N from STATE.md
|
|
14
|
+
- Warn if `steps_complete` is empty (no steps executed yet)
|
|
15
|
+
|
|
16
|
+
2. **Gather scope.**
|
|
17
|
+
- Collect changed files via `git diff --name-only HEAD`
|
|
18
|
+
- If no changes, use all files in the current phase directory
|
|
19
|
+
- Run `codegraph action=check` — if available, use `codegraph_impact` on changed files to surface dependent modules not caught by `git diff`
|
|
20
|
+
|
|
21
|
+
3. **Run four checks in parallel:**
|
|
22
|
+
|
|
23
|
+
- **Tests (@tester):** `npm test` — all tests must pass, no failures or unexplained skips
|
|
24
|
+
|
|
25
|
+
- **Code Review (@reviewer):** Review all changed files — security (secrets, injection, auth gaps), quality (critical bugs, error handling, TDD discipline), conventions (naming, patterns, import style), >= 80% test coverage for changed files. If UI-heavy: design fidelity review against approved design artifact
|
|
26
|
+
|
|
27
|
+
- **UI Design Review (@design):** If UI-heavy — compare implemented UI to approved design artifact, report on hierarchy, spacing, CTA flow, responsiveness, accessibility, and missing state coverage. Fail verification on severe design fidelity mismatch
|
|
28
|
+
|
|
29
|
+
- **Security Scan (@security-auditor):** No hardcoded secrets, input validation at trust boundaries, auth/authz on all protected routes, no CRITICAL/HIGH vulnerabilities
|
|
30
|
+
|
|
31
|
+
- **Deploy Check:** `npm audit --audit-level=high` and `npm run build` — no HIGH/CRITICAL CVEs, build must succeed
|
|
32
|
+
|
|
33
|
+
4. **Aggregate results.** Present consolidated table with pass/fail status for each check.
|
|
34
|
+
|
|
35
|
+
5. **Go/No-Go decision.**
|
|
36
|
+
|
|
37
|
+
**VERIFIED (all checks pass):**
|
|
38
|
+
- Update STATE.md: `status: verified`, `last_action: "Phase N verified — all checks passed"`, `verified_at: <timestamp>`
|
|
39
|
+
- Report next steps (deploy, increment phase, etc.)
|
|
40
|
+
|
|
41
|
+
**NOT VERIFIED (one or more checks fail):**
|
|
42
|
+
- List required fixes
|
|
43
|
+
- Do NOT update STATE.md to verified status
|
|
44
|
+
- Report suggested next step (run `/fd-execute` for fixes, then `/fd-verify` again)
|
|
45
|
+
|
|
46
|
+
6. **No-Go conditions (automatic NOT VERIFIED):** test failures, CRITICAL security vulnerability, unpatched HIGH/CRITICAL CVE, build error, CRITICAL code review finding.
|
|
47
|
+
|
|
48
|
+
## Output / State
|
|
49
|
+
|
|
50
|
+
STATE.md on verified:
|
|
51
|
+
```yaml
|
|
52
|
+
status: verified
|
|
53
|
+
last_action: "Phase N verified — all checks passed"
|
|
54
|
+
verified_at: "<timestamp>"
|
|
55
|
+
```
|
|
56
|
+
|
|
57
|
+
## Examples
|
|
58
|
+
|
|
59
|
+
```
|
|
60
|
+
/fd-verify
|
|
61
|
+
```
|
|
62
|
+
|
|
63
|
+
Run full verification pipeline for the current phase.
|
|
64
|
+
|
|
65
|
+
```
|
|
66
|
+
/fd-verify --phase=2 --env=staging
|
|
67
|
+
```
|
|
68
|
+
|
|
69
|
+
Verify phase 2 and run deploy check against staging environment.
|
|
70
|
+
|
|
71
|
+
## Related Commands
|
|
72
|
+
|
|
73
|
+
- `/fd-execute` — implement the feature before verification
|
|
74
|
+
- `/fd-plan` — create the plan that was verified against
|
|
75
|
+
- `/fd-resume` — reload state after making fixes
|