sdd-flow 1.0.0
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/LICENSE +21 -0
- package/README.md +83 -0
- package/bin/cli.js +315 -0
- package/package.json +32 -0
- package/templates/AGENTS.md +49 -0
- package/templates/QUICKSTART.md +50 -0
- package/templates/agents/AGENTS.md +92 -0
- package/templates/agents/DECISIONS.md +21 -0
- package/templates/agents/PROJECT.md +145 -0
- package/templates/agents/RULES.md +116 -0
- package/templates/agents/SETUP.md +224 -0
- package/templates/agents/harness/claude-code/.hooks/guard-delete.sh +22 -0
- package/templates/agents/harness/claude-code/.hooks/guard-deps.sh +25 -0
- package/templates/agents/harness/claude-code/.hooks/guard-git.sh +24 -0
- package/templates/agents/harness/claude-code/commands/sdd-implement.md +3 -0
- package/templates/agents/harness/claude-code/commands/sdd-plan.md +3 -0
- package/templates/agents/harness/claude-code/commands/sdd-research.md +3 -0
- package/templates/agents/harness/claude-code/commands/sdd-validate.md +3 -0
- package/templates/agents/harness/claude-code/settings.json +34 -0
- package/templates/agents/harness/opencode/commands/sdd-implement.md +3 -0
- package/templates/agents/harness/opencode/commands/sdd-plan.md +3 -0
- package/templates/agents/harness/opencode/commands/sdd-research.md +3 -0
- package/templates/agents/harness/opencode/commands/sdd-validate.md +3 -0
- package/templates/agents/harness/opencode/opencode.json +18 -0
- package/templates/agents/prompts/conventional-commit.md +51 -0
- package/templates/agents/prompts/pr-template.md +21 -0
- package/templates/agents/prompts/rpi-implement.md +46 -0
- package/templates/agents/prompts/rpi-plan.md +117 -0
- package/templates/agents/prompts/rpi-research.md +67 -0
- package/templates/agents/prompts/rpi-validate.md +96 -0
- package/templates/agents/prompts/skill-call.md +29 -0
- package/templates/agents/prompts/task-create.md +42 -0
- package/templates/agents/skills/_template/SKILL.md +37 -0
|
@@ -0,0 +1,51 @@
|
|
|
1
|
+
# Prompt: Conventional Commit Suggestion
|
|
2
|
+
|
|
3
|
+
Based on the changes made in this task/feature, suggest commit messages following the Conventional Commits specification.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## Rules
|
|
8
|
+
|
|
9
|
+
- Do not run git commands
|
|
10
|
+
- Generate 2 suggestions:
|
|
11
|
+
1. In the team's primary language
|
|
12
|
+
2. In English
|
|
13
|
+
- Use a scope when it makes sense (e.g. auth, api, ui, db, config, setup)
|
|
14
|
+
- Include a short body with bullet points if the change warrants it
|
|
15
|
+
|
|
16
|
+
## Output format
|
|
17
|
+
|
|
18
|
+
**[Primary language]:**
|
|
19
|
+
`<message>`
|
|
20
|
+
|
|
21
|
+
**EN:**
|
|
22
|
+
`<message>`
|
|
23
|
+
|
|
24
|
+
---
|
|
25
|
+
|
|
26
|
+
## Commit Types
|
|
27
|
+
|
|
28
|
+
| Type | When to use | SemVer impact |
|
|
29
|
+
|---|---|---|
|
|
30
|
+
| `feat` | New feature or capability | minor |
|
|
31
|
+
| `fix` | Bug fix | patch |
|
|
32
|
+
| `refactor` | Code change with no behavior change | none |
|
|
33
|
+
| `perf` | Performance improvement | none |
|
|
34
|
+
| `docs` | Documentation only | none |
|
|
35
|
+
| `style` | Formatting, whitespace (no logic change) | none |
|
|
36
|
+
| `test` | Adding or fixing tests | none |
|
|
37
|
+
| `build` | Build system or external dependency changes | none |
|
|
38
|
+
| `ci` | CI/CD configuration changes | none |
|
|
39
|
+
| `chore` | Maintenance tasks (no src or test changes) | none |
|
|
40
|
+
| `revert` | Reverts a previous commit | depends |
|
|
41
|
+
|
|
42
|
+
Use `!` suffix (e.g. `feat!:`) to indicate a breaking change (major in SemVer).
|
|
43
|
+
|
|
44
|
+
## Examples
|
|
45
|
+
|
|
46
|
+
```
|
|
47
|
+
feat(auth): add Google OAuth login
|
|
48
|
+
fix(api): correct JSON parsing error in response handler
|
|
49
|
+
refactor(db): replace raw queries with repository pattern
|
|
50
|
+
feat!: change authentication token format (breaking change)
|
|
51
|
+
```
|
|
@@ -0,0 +1,46 @@
|
|
|
1
|
+
# Prompt: Implement (RPI — Phase 3)
|
|
2
|
+
|
|
3
|
+
Use this prompt to implement a single task from an existing spec.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## Agent instruction
|
|
8
|
+
|
|
9
|
+
You will implement **one single task** from an already planned feature.
|
|
10
|
+
|
|
11
|
+
**Before starting, read:**
|
|
12
|
+
- `agents/AGENTS.md`
|
|
13
|
+
- `agents/PROJECT.md`
|
|
14
|
+
- `agents/RULES.md`
|
|
15
|
+
- `agents/specs/NNN-name/SPEC.md`
|
|
16
|
+
- `agents/specs/NNN-name/TASK.md` — identify the first unchecked `[ ]`
|
|
17
|
+
- `agents/specs/NNN-name/PROGRESS.md`
|
|
18
|
+
- Run `grep -A 8 "## When to use" agents/skills/*/SKILL.md` — load the full `SKILL.md` only for skills relevant to this task
|
|
19
|
+
|
|
20
|
+
**Expected input:**
|
|
21
|
+
- Spec number (e.g. `001`)
|
|
22
|
+
- Task to implement (e.g. `T03`) — if not provided, use the first unchecked `[ ]`
|
|
23
|
+
|
|
24
|
+
---
|
|
25
|
+
|
|
26
|
+
## Expected behavior
|
|
27
|
+
|
|
28
|
+
1. Declare which task will be implemented before generating any code
|
|
29
|
+
2. Implement only what is necessary for that single task
|
|
30
|
+
3. State exactly **which file** was created or modified
|
|
31
|
+
4. After completing:
|
|
32
|
+
- Confirm the `verify:` condition on the completed task is met before marking it done
|
|
33
|
+
- Mark `[x]` on the task in TASK.md
|
|
34
|
+
- Update PROGRESS.md with what was done and any observations
|
|
35
|
+
- Record relevant architectural decisions in DECISIONS.md
|
|
36
|
+
5. **Stop and wait for confirmation** before moving to the next task
|
|
37
|
+
|
|
38
|
+
---
|
|
39
|
+
|
|
40
|
+
## Constraints
|
|
41
|
+
|
|
42
|
+
- Do not anticipate future tasks
|
|
43
|
+
- Do not refactor code outside the current task's scope
|
|
44
|
+
- Do not add dependencies without recording them in DECISIONS.md
|
|
45
|
+
- Do not run git commands
|
|
46
|
+
- If there is ambiguity about scope or approach, **ask before generating code**
|
|
@@ -0,0 +1,117 @@
|
|
|
1
|
+
# Prompt: Plan (RPI — Phase 2)
|
|
2
|
+
|
|
3
|
+
Use this prompt to generate all planning artifacts for a spec.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## Agent instruction
|
|
8
|
+
|
|
9
|
+
You will plan the implementation of an already-researched feature.
|
|
10
|
+
|
|
11
|
+
**Before starting, read:**
|
|
12
|
+
- `agents/AGENTS.md`
|
|
13
|
+
- `agents/PROJECT.md`
|
|
14
|
+
- `agents/RULES.md`
|
|
15
|
+
- `agents/specs/NNN-name/RESEARCH.md`
|
|
16
|
+
- Check which files already exist in `agents/specs/NNN-name/`
|
|
17
|
+
|
|
18
|
+
**Expected input:**
|
|
19
|
+
- Spec number and name (e.g. `001-user-authentication`)
|
|
20
|
+
|
|
21
|
+
---
|
|
22
|
+
|
|
23
|
+
## Adaptive behavior
|
|
24
|
+
|
|
25
|
+
| Situation | Action |
|
|
26
|
+
|---|---|
|
|
27
|
+
| SPEC.md does not exist | Create all artifacts from scratch |
|
|
28
|
+
| SPEC.md exists, TASK.md does not | Generate only TASK.md and PROGRESS.md based on existing SPEC.md |
|
|
29
|
+
| All artifacts exist | Review and update only what is incomplete or outdated |
|
|
30
|
+
|
|
31
|
+
---
|
|
32
|
+
|
|
33
|
+
## Deliverables
|
|
34
|
+
|
|
35
|
+
Generate the following files inside `agents/specs/NNN-name/`:
|
|
36
|
+
|
|
37
|
+
### SPEC.md
|
|
38
|
+
```markdown
|
|
39
|
+
# SPEC — [Feature Name]
|
|
40
|
+
|
|
41
|
+
## Goal
|
|
42
|
+
What this feature delivers (refined from RESEARCH.md).
|
|
43
|
+
|
|
44
|
+
## Entities involved
|
|
45
|
+
Data structures created or modified, with full field list.
|
|
46
|
+
|
|
47
|
+
## Business Rules
|
|
48
|
+
Numbered list. Only rules that directly affect implementation.
|
|
49
|
+
|
|
50
|
+
## Expected behaviors
|
|
51
|
+
Per behavior: what triggers it, what it does, what the output is.
|
|
52
|
+
|
|
53
|
+
## UI
|
|
54
|
+
Describe components, states, interactions and validations.
|
|
55
|
+
Write `N/A` if no UI is involved.
|
|
56
|
+
|
|
57
|
+
## Out of scope
|
|
58
|
+
Explicit list of what this spec does NOT cover.
|
|
59
|
+
```
|
|
60
|
+
|
|
61
|
+
### TASK.md
|
|
62
|
+
```markdown
|
|
63
|
+
# TASK — [Feature Name]
|
|
64
|
+
|
|
65
|
+
## Tasks
|
|
66
|
+
|
|
67
|
+
- [ ] T01 — [Verb + object + target file] (e.g. Create User entity schema in config)
|
|
68
|
+
verify: [observable condition confirming completion — e.g. endpoint returns 201 with `id` field; returns 400 if `email` is missing]
|
|
69
|
+
- [ ] T02 — ...
|
|
70
|
+
verify: [specific and testable condition — never "works correctly"]
|
|
71
|
+
```
|
|
72
|
+
|
|
73
|
+
Task rules:
|
|
74
|
+
- One task = one file created/modified, or one isolated function
|
|
75
|
+
- Order must respect dependencies (schema → data layer → service → route → UI)
|
|
76
|
+
- No circular dependencies between tasks
|
|
77
|
+
- Preferred verbs: `Add`, `Create`, `Implement`, `Update`, `Extract`, `Register`
|
|
78
|
+
- Each task must include a `verify:` line with a specific, observable condition — never "works correctly"
|
|
79
|
+
|
|
80
|
+
### PROGRESS.md
|
|
81
|
+
```markdown
|
|
82
|
+
# PROGRESS — [Feature Name]
|
|
83
|
+
|
|
84
|
+
## Status: 🔴 Not started
|
|
85
|
+
|
|
86
|
+
## Tasks
|
|
87
|
+
|
|
88
|
+
| Task | Status | Notes |
|
|
89
|
+
|---|---|---|
|
|
90
|
+
| T01 | Pending | |
|
|
91
|
+
| T02 | Pending | |
|
|
92
|
+
|
|
93
|
+
## History
|
|
94
|
+
|
|
95
|
+
(updated during implementation)
|
|
96
|
+
```
|
|
97
|
+
|
|
98
|
+
### TEST.md
|
|
99
|
+
```markdown
|
|
100
|
+
# TEST — [Feature Name]
|
|
101
|
+
|
|
102
|
+
## Acceptance Test Cases
|
|
103
|
+
|
|
104
|
+
### TC01 — [Behavior being tested]
|
|
105
|
+
- **Given** [initial state]
|
|
106
|
+
- **When** [action performed]
|
|
107
|
+
- **Then** [expected outcome]
|
|
108
|
+
```
|
|
109
|
+
|
|
110
|
+
---
|
|
111
|
+
|
|
112
|
+
## Constraints
|
|
113
|
+
|
|
114
|
+
- Do not write code in this phase
|
|
115
|
+
- Do not implement any task — only plan
|
|
116
|
+
- If the RESEARCH.md has unresolved open questions, **stop and ask** before planning
|
|
117
|
+
- After delivering all artifacts, **wait for approval** before advancing to Implement
|
|
@@ -0,0 +1,67 @@
|
|
|
1
|
+
# Prompt: Research (RPI — Phase 1)
|
|
2
|
+
|
|
3
|
+
Use this prompt to analyze the scope of a feature and produce RESEARCH.md.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## Agent instruction
|
|
8
|
+
|
|
9
|
+
You will analyze the scope of a feature and produce the `RESEARCH.md` for the spec.
|
|
10
|
+
This file is the only artifact of this phase and will be the context for the Plan phase.
|
|
11
|
+
|
|
12
|
+
**Before starting, read:**
|
|
13
|
+
- `agents/AGENTS.md`
|
|
14
|
+
- `agents/PROJECT.md`
|
|
15
|
+
- `agents/RULES.md`
|
|
16
|
+
|
|
17
|
+
**Expected input:**
|
|
18
|
+
- Requirements source (PRD, textual description, or any document provided)
|
|
19
|
+
- Spec name and number (e.g. `001-user-authentication`)
|
|
20
|
+
|
|
21
|
+
---
|
|
22
|
+
|
|
23
|
+
## Deliverable: RESEARCH.md
|
|
24
|
+
|
|
25
|
+
Produce **only** the file `agents/specs/NNN-name/RESEARCH.md` with the structure below.
|
|
26
|
+
Be concise — this file will be read in future sessions as compressed context.
|
|
27
|
+
|
|
28
|
+
```markdown
|
|
29
|
+
# RESEARCH — [Feature Name]
|
|
30
|
+
|
|
31
|
+
## Goal
|
|
32
|
+
One clear sentence stating what this feature delivers.
|
|
33
|
+
|
|
34
|
+
## Entities and data structures
|
|
35
|
+
List the entities involved, data structures to be created or modified,
|
|
36
|
+
and their expected fields/schema.
|
|
37
|
+
|
|
38
|
+
## Relevant business rules
|
|
39
|
+
Only the rules that directly impact implementation.
|
|
40
|
+
Reference by number if the PRD uses numbered rules.
|
|
41
|
+
|
|
42
|
+
## Dependencies
|
|
43
|
+
Features or specs that must be complete before this one.
|
|
44
|
+
List `None` if there are none.
|
|
45
|
+
|
|
46
|
+
## Scope decisions
|
|
47
|
+
What is explicitly inside and outside the scope of this spec.
|
|
48
|
+
Be explicit about edge cases.
|
|
49
|
+
|
|
50
|
+
## Risks and attention points
|
|
51
|
+
Resolved ambiguities, identified trade-offs, and warnings for the implementation phase.
|
|
52
|
+
|
|
53
|
+
## Open questions
|
|
54
|
+
List only questions that block the Plan phase.
|
|
55
|
+
If none, write `None`.
|
|
56
|
+
```
|
|
57
|
+
|
|
58
|
+
---
|
|
59
|
+
|
|
60
|
+
## Constraints
|
|
61
|
+
|
|
62
|
+
- Produce only RESEARCH.md — no other file must be created
|
|
63
|
+
- Do not create SPEC.md, TASK.md, PROGRESS.md, TEST.md or any code file
|
|
64
|
+
- Do not write code in this phase
|
|
65
|
+
- Do not anticipate implementation decisions
|
|
66
|
+
- If there are open questions, **stop and wait for answers** before delivering RESEARCH.md
|
|
67
|
+
- After delivering RESEARCH.md, **wait for approval** before advancing to Plan
|
|
@@ -0,0 +1,96 @@
|
|
|
1
|
+
# Prompt: Validate (RPI — Phase 4)
|
|
2
|
+
|
|
3
|
+
Use this prompt to verify that the implementation satisfies the spec and all acceptance criteria.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## Agent instruction
|
|
8
|
+
|
|
9
|
+
You will validate the implementation of a completed feature against its spec and test cases.
|
|
10
|
+
|
|
11
|
+
**Before starting, read:**
|
|
12
|
+
- `agents/AGENTS.md`
|
|
13
|
+
- `agents/PROJECT.md`
|
|
14
|
+
- `agents/RULES.md`
|
|
15
|
+
- `agents/specs/NNN-name/SPEC.md`
|
|
16
|
+
- `agents/specs/NNN-name/TASK.md` — confirm all tasks are marked `[x]`
|
|
17
|
+
- `agents/specs/NNN-name/TEST.md`
|
|
18
|
+
- `agents/specs/NNN-name/PROGRESS.md`
|
|
19
|
+
- The code files produced during Implement (listed in PROGRESS.md History)
|
|
20
|
+
|
|
21
|
+
**Expected input:**
|
|
22
|
+
- Spec number and name (e.g. `001-user-authentication`)
|
|
23
|
+
|
|
24
|
+
---
|
|
25
|
+
|
|
26
|
+
## Validation procedure
|
|
27
|
+
|
|
28
|
+
### Step 1 — Confirm implementation is complete
|
|
29
|
+
Check that every task in `TASK.md` is marked `[x]`.
|
|
30
|
+
If any `[ ]` remains, **stop and report** — do not proceed with validation.
|
|
31
|
+
|
|
32
|
+
### Step 2 — Verify task-level conditions
|
|
33
|
+
For each task in `TASK.md`, read its `verify:` condition and check whether
|
|
34
|
+
the produced code satisfies it. Examine the actual files — do not assume.
|
|
35
|
+
|
|
36
|
+
### Step 3 — Verify acceptance test cases
|
|
37
|
+
For each test case in `TEST.md`, assess whether the implementation satisfies
|
|
38
|
+
the Given/When/Then scenario. Read the relevant code directly.
|
|
39
|
+
|
|
40
|
+
### Step 4 — Produce VALIDATION.md
|
|
41
|
+
Write `agents/specs/NNN-name/VALIDATION.md` with the structure below.
|
|
42
|
+
|
|
43
|
+
---
|
|
44
|
+
|
|
45
|
+
## Deliverable: VALIDATION.md
|
|
46
|
+
|
|
47
|
+
```markdown
|
|
48
|
+
# VALIDATION — [Feature Name]
|
|
49
|
+
|
|
50
|
+
## Status: [✅ Approved | ⚠️ Partial | ❌ Failed]
|
|
51
|
+
|
|
52
|
+
---
|
|
53
|
+
|
|
54
|
+
## Task Verification
|
|
55
|
+
|
|
56
|
+
| Task | verify: condition | Status |
|
|
57
|
+
|---|---|---|
|
|
58
|
+
| T01 | [condition from TASK.md] | ✅ Met / ❌ Not met / ⚠️ Partial |
|
|
59
|
+
| T02 | ... | ... |
|
|
60
|
+
|
|
61
|
+
---
|
|
62
|
+
|
|
63
|
+
## Acceptance Test Coverage
|
|
64
|
+
|
|
65
|
+
| Test Case | Description | Status | Evidence |
|
|
66
|
+
|---|---|---|---|
|
|
67
|
+
| TC01 | [behavior being tested] | ✅ Verified / ❌ Not met / ⚠️ Partial | [file and line or behavior observed in code] |
|
|
68
|
+
| TC02 | ... | ... | ... |
|
|
69
|
+
|
|
70
|
+
---
|
|
71
|
+
|
|
72
|
+
## Gaps Found
|
|
73
|
+
|
|
74
|
+
List any `verify:` conditions or test cases that are not fully met.
|
|
75
|
+
Write `None` if everything is satisfied.
|
|
76
|
+
|
|
77
|
+
- [ ] [Gap description — what is missing or incorrect]
|
|
78
|
+
|
|
79
|
+
---
|
|
80
|
+
|
|
81
|
+
## Recommendation
|
|
82
|
+
|
|
83
|
+
[Approve — all criteria met]
|
|
84
|
+
[Request fixes — list the gaps that must be resolved before approval]
|
|
85
|
+
```
|
|
86
|
+
|
|
87
|
+
---
|
|
88
|
+
|
|
89
|
+
## Constraints
|
|
90
|
+
|
|
91
|
+
- Do not write or modify any code in this phase
|
|
92
|
+
- Do not mark gaps as acceptable — report them, even if minor
|
|
93
|
+
- If a test case cannot be verified from static code analysis alone, flag it explicitly:
|
|
94
|
+
`⚠️ Requires runtime verification — cannot confirm from code alone`
|
|
95
|
+
- After delivering VALIDATION.md, **wait for developer review**
|
|
96
|
+
- If gaps are found and fixed, a new Validate session must be run — do not reuse this report
|
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
# Prompt: Execute task using a Skill
|
|
2
|
+
|
|
3
|
+
Use this prompt when a task requires following a specific skill pattern.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## Agent instruction
|
|
8
|
+
|
|
9
|
+
Read:
|
|
10
|
+
- `agents/AGENTS.md`
|
|
11
|
+
- `agents/PROJECT.md`
|
|
12
|
+
- `agents/RULES.md`
|
|
13
|
+
- `agents/specs/<SPEC_NUMBER>/SPEC.md`
|
|
14
|
+
- `agents/specs/<SPEC_NUMBER>/TASK.md`
|
|
15
|
+
- `agents/specs/<SPEC_NUMBER>/PROGRESS.md`
|
|
16
|
+
- `agents/skills/<SKILL_NAME>/SKILL.md`
|
|
17
|
+
|
|
18
|
+
Now implement ONLY the task: `<TASK_ID>`
|
|
19
|
+
|
|
20
|
+
Use the skill as the mandatory pattern source for this task.
|
|
21
|
+
|
|
22
|
+
---
|
|
23
|
+
|
|
24
|
+
## Rules
|
|
25
|
+
|
|
26
|
+
- Do not run git commands
|
|
27
|
+
- Do not add comments or documentation unless explicitly required by the skill
|
|
28
|
+
- Update PROGRESS.md upon completion
|
|
29
|
+
- Stop and wait for confirmation before the next task
|
|
@@ -0,0 +1,42 @@
|
|
|
1
|
+
# Prompt: Task Create
|
|
2
|
+
|
|
3
|
+
Use this prompt to create or revise the TASK.md of an existing spec.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## Agent instruction
|
|
8
|
+
|
|
9
|
+
From an already approved SPEC.md, generate or update the TASK.md with a proper task breakdown.
|
|
10
|
+
|
|
11
|
+
**Expected input:**
|
|
12
|
+
- Content or path of the SPEC.md
|
|
13
|
+
|
|
14
|
+
---
|
|
15
|
+
|
|
16
|
+
## Task generation rules
|
|
17
|
+
|
|
18
|
+
- One task = one file created or modified, or one isolated function
|
|
19
|
+
- Order must respect dependencies:
|
|
20
|
+
- Schema / config → data access → business logic → routing → UI
|
|
21
|
+
- Tasks that are consumed by other tasks must come first
|
|
22
|
+
- No task should have a circular dependency on another
|
|
23
|
+
- Each task must be independently reviewable and verifiable
|
|
24
|
+
|
|
25
|
+
## Required format
|
|
26
|
+
|
|
27
|
+
```
|
|
28
|
+
# TASK — [Feature Name]
|
|
29
|
+
|
|
30
|
+
## Tasks
|
|
31
|
+
|
|
32
|
+
- [ ] T01 — [Verb + object + target file] (e.g. Add User schema to config module)
|
|
33
|
+
- [ ] T02 — ...
|
|
34
|
+
```
|
|
35
|
+
|
|
36
|
+
## Preferred verbs
|
|
37
|
+
|
|
38
|
+
`Add`, `Create`, `Implement`, `Update`, `Extract`, `Register`, `Remove`, `Move`
|
|
39
|
+
|
|
40
|
+
---
|
|
41
|
+
|
|
42
|
+
**After generating TASK.md, stop and wait for approval before implementing.**
|
|
@@ -0,0 +1,37 @@
|
|
|
1
|
+
# Skill: [Capability Name]
|
|
2
|
+
|
|
3
|
+
## When to use this skill
|
|
4
|
+
|
|
5
|
+
[Specific task types that trigger this skill. Be precise — the agent uses this section
|
|
6
|
+
to decide whether to load the skill. Vague descriptions reduce detection accuracy.]
|
|
7
|
+
|
|
8
|
+
---
|
|
9
|
+
|
|
10
|
+
## Patterns
|
|
11
|
+
|
|
12
|
+
[Correct approach with code examples from this project's stack]
|
|
13
|
+
|
|
14
|
+
```language
|
|
15
|
+
// example
|
|
16
|
+
```
|
|
17
|
+
|
|
18
|
+
---
|
|
19
|
+
|
|
20
|
+
## Anti-patterns
|
|
21
|
+
|
|
22
|
+
| Anti-pattern | Why to avoid |
|
|
23
|
+
|---|---|
|
|
24
|
+
| [What NOT to do] | [Reason] |
|
|
25
|
+
|
|
26
|
+
---
|
|
27
|
+
|
|
28
|
+
## Constraints
|
|
29
|
+
|
|
30
|
+
- [Hard rule specific to this capability]
|
|
31
|
+
|
|
32
|
+
---
|
|
33
|
+
|
|
34
|
+
## Checklist
|
|
35
|
+
|
|
36
|
+
- [ ] [Item the agent must verify before finishing a task using this skill]
|
|
37
|
+
- [ ] [Another verification item]
|