kiro-review

Name: kiro-review
Author: gotalab

3.6kstars271forks

How to install kiro-review?

gotalab/cc-sdd/kiro-review

$npx -y skills add gotalab/cc-sdd --skill kiro-review

Installs into the current project.

Summary

Review a task implementation against approved specs, task boundaries, and verification evidence. Use after an implementer finishes a task, after remediation, or before accepting a task as complete.

Files · 1

View on GitHub

SKILL.md

1	# kiro-review
2
3	<background_information>
4	This skill performs task-local adversarial review. It verifies that the implementation is real, complete, bounded, aligned with approved requirements and design, and supported by mechanical verification evidence.
5
6	Boundary terminology continuity:
7	- discovery identifies `Boundary Candidates`
8	- design fixes `Boundary Commitments`
9	- tasks constrain execution with `_Boundary:_`
10	- review rejects concrete `Boundary Violations`
11	</background_information>
12
13	<instructions>
14	## When to Use
15
16	- After an implementer reports `READY_FOR_REVIEW`
17	- After remediation for a rejected review
18	- Before marking a task `[x]`
19	- Before accepting a task into feature-level validation
20
21	Do not use this skill to invent missing requirements or silently reinterpret the spec.
22
23	## Inputs
24
25	Provide:
26	- Task ID and exact task text from `tasks.md`
27	- Relevant requirement section numbers
28	- Relevant design section numbers
29	- Spec file paths (`requirements.md`, `design.md`, optionally `tasks.md`)
30	- The implementer's status report
31	- The task `_Boundary:_` scope constraints
32	- Validation commands discovered by the controller
33	- Relevant steering excerpts when applicable
34	- Relevant `## Implementation Notes` entries when applicable
35
36	## Outputs
37
38	Return one of:
39	- `APPROVED`
40	- `REJECTED`
41
42	Also return:
43	- Mechanical results
44	- Findings with severity
45	- Required remediation
46	- One-sentence summary
47
48	Use the language specified in `spec.json`.
49
50	## First Action
51
52	Run `git diff` to inspect the actual code changes. If the diff is large or ambiguous, read the changed files directly. Do not trust the implementer report as source of truth.
53
54	## Core Principle
55
56	Read the spec yourself. Read the diff yourself. Verify mechanically where possible. Reject on concrete failures rather than interpretive optimism.
57	The main review question is not just "does it work?" but "does it stay inside the approved responsibility boundary without hiding new coupling?"
58
59	## Mechanical Checks
60
61	Run these checks and use the result as primary signal.
62
63	### 1. Regression Safety
64	- Run the project's canonical test suite using the validation commands discovered by the controller.
65	- If tests fail, reject.
66
67	### 2. No Residual Placeholder Markers
68	- Check changed files for `TBD`, `TODO`, `FIXME`, `HACK`, `XXX`.
69	- Reject if new placeholder markers were introduced without explicit task justification.
70
71	### 3. No Hardcoded Secrets
72	- Check changed files for hardcoded secrets or credentials.
73	- Reject if concrete secret patterns are introduced.
74
75	### 4. Boundary Respect
76	- Compare changed files against the task `_Boundary:_` scope.
77	- Reject if the change spills outside the approved boundary without explicit justification.
78	- Reject if the implementation introduces hidden cross-boundary coordination inside what should be a local task.
79
80	### 5. RED Phase Evidence
81	- For behavioral tasks, verify that the implementer status report includes `RED_PHASE_OUTPUT`.
82	- Reject if RED evidence is missing, empty, or unrelated to the task's acceptance criteria.
83
84	### 6. Runtime-Sensitive Static Checks
85	- If the project already has lint or equivalent static analysis for the touched stack, run the relevant command for the task boundary.
86	- Pay attention to patterns that can survive typecheck/build yet fail at runtime: type-only imports used as values, missing namespace value imports for qualified-name access, unresolved globals, and newly introduced runtime-sensitive dependencies without matching boot/runtime handling.
87	- If no project lint command exists, perform a targeted diff-based spot check in the changed files for those patterns.
88	- Reject on concrete findings that create a realistic boot-time or module-load failure.
89
90	## Judgment Checks
91
92	### 7. Reality Check
93	- Confirm the implementation is real production code, not a placeholder, stub, fake path, or deferred-work shell.
94
95	### 8. Acceptance Criteria Coverage
96	- Read the task description and confirm all aspects are implemented, not only the primary happy path.
97
98	### 9. Requirements Alignment
99	- Read the referenced sections in `requirements.md`.
100	- Confirm each requirement is satisfied by concrete observable behavior.
101	- Use original section numbers only.
102
103	### 10. Design Alignment
104	- Read the referenced sections in `design.md`.
105	- Confirm the implementation uses the prescribed structures, interfaces, and dependency direction.
106	- Reject silent substitutions for design-mandated choices.
107
108	### 10.5 Boundary Audit
109	- Compare the implementation against the design's boundary commitments and out-of-boundary statements.
110	- Reject if downstream-specific behavior is pushed into an upstream boundary for convenience.
111	- Reject if the implementation creates new hidden dependencies, shared ownership, or undeclared coupling across adjacent boundaries.

People who liked this also viewed

6 picks

Type

requesting-code-reviewUse when completing tasks, implementing major features, or before merging to verify work meets requirementsSkillsJul 2026176k259k
code-reviewSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026158k181k
receiving-code-reviewUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performativeSkillsJul 2026146k259k
baoyu-format-markdownFormats plain text or markdown files with frontmatter, titles, summaries, headings, bold, lists, and code blocks. Use when user asks to "format markdown", "beautify article", "add formatting", or imprSkillsJul 202625k24k
code-review-excellenceMaster effective code review practices to provide constructive feedback, catch bugs early, and foster knowledge sharing while maintaining team morale. Use when reviewing pull requests, establishing reSkillsJul 202625k38k
code-review-and-qualityConducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensiSkillsJul 202618k80k

kiro-review

3.6kstars271forks

View on GitHub

How to install kiro-review?

gotalab/cc-sdd/kiro-review

$npx -y skills add gotalab/cc-sdd --skill kiro-review

Installs into the current project.

Summary

Review a task implementation against approved specs, task boundaries, and verification evidence. Use after an implementer finishes a task, after remediation, or before accepting a task as complete.

Files · 1

View on GitHub

SKILL.md

1	# kiro-review
2
3	<background_information>
4	This skill performs task-local adversarial review. It verifies that the implementation is real, complete, bounded, aligned with approved requirements and design, and supported by mechanical verification evidence.
5
6	Boundary terminology continuity:
7	- discovery identifies `Boundary Candidates`
8	- design fixes `Boundary Commitments`
9	- tasks constrain execution with `_Boundary:_`
10	- review rejects concrete `Boundary Violations`
11	</background_information>
12
13	<instructions>
14	## When to Use
15
16	- After an implementer reports `READY_FOR_REVIEW`
17	- After remediation for a rejected review
18	- Before marking a task `[x]`
19	- Before accepting a task into feature-level validation
20
21	Do not use this skill to invent missing requirements or silently reinterpret the spec.
22
23	## Inputs
24
25	Provide:
26	- Task ID and exact task text from `tasks.md`
27	- Relevant requirement section numbers
28	- Relevant design section numbers
29	- Spec file paths (`requirements.md`, `design.md`, optionally `tasks.md`)
30	- The implementer's status report
31	- The task `_Boundary:_` scope constraints
32	- Validation commands discovered by the controller
33	- Relevant steering excerpts when applicable
34	- Relevant `## Implementation Notes` entries when applicable
35
36	## Outputs
37
38	Return one of:
39	- `APPROVED`
40	- `REJECTED`
41
42	Also return:
43	- Mechanical results
44	- Findings with severity
45	- Required remediation
46	- One-sentence summary
47
48	Use the language specified in `spec.json`.
49
50	## First Action
51
52	Run `git diff` to inspect the actual code changes. If the diff is large or ambiguous, read the changed files directly. Do not trust the implementer report as source of truth.
53
54	## Core Principle
55
56	Read the spec yourself. Read the diff yourself. Verify mechanically where possible. Reject on concrete failures rather than interpretive optimism.
57	The main review question is not just "does it work?" but "does it stay inside the approved responsibility boundary without hiding new coupling?"
58
59	## Mechanical Checks
60
61	Run these checks and use the result as primary signal.
62
63	### 1. Regression Safety
64	- Run the project's canonical test suite using the validation commands discovered by the controller.
65	- If tests fail, reject.
66
67	### 2. No Residual Placeholder Markers
68	- Check changed files for `TBD`, `TODO`, `FIXME`, `HACK`, `XXX`.
69	- Reject if new placeholder markers were introduced without explicit task justification.
70
71	### 3. No Hardcoded Secrets
72	- Check changed files for hardcoded secrets or credentials.
73	- Reject if concrete secret patterns are introduced.
74
75	### 4. Boundary Respect
76	- Compare changed files against the task `_Boundary:_` scope.
77	- Reject if the change spills outside the approved boundary without explicit justification.
78	- Reject if the implementation introduces hidden cross-boundary coordination inside what should be a local task.
79
80	### 5. RED Phase Evidence
81	- For behavioral tasks, verify that the implementer status report includes `RED_PHASE_OUTPUT`.
82	- Reject if RED evidence is missing, empty, or unrelated to the task's acceptance criteria.
83
84	### 6. Runtime-Sensitive Static Checks
85	- If the project already has lint or equivalent static analysis for the touched stack, run the relevant command for the task boundary.
86	- Pay attention to patterns that can survive typecheck/build yet fail at runtime: type-only imports used as values, missing namespace value imports for qualified-name access, unresolved globals, and newly introduced runtime-sensitive dependencies without matching boot/runtime handling.
87	- If no project lint command exists, perform a targeted diff-based spot check in the changed files for those patterns.
88	- Reject on concrete findings that create a realistic boot-time or module-load failure.
89
90	## Judgment Checks
91
92	### 7. Reality Check
93	- Confirm the implementation is real production code, not a placeholder, stub, fake path, or deferred-work shell.
94
95	### 8. Acceptance Criteria Coverage
96	- Read the task description and confirm all aspects are implemented, not only the primary happy path.
97
98	### 9. Requirements Alignment
99	- Read the referenced sections in `requirements.md`.
100	- Confirm each requirement is satisfied by concrete observable behavior.
101	- Use original section numbers only.
102
103	### 10. Design Alignment
104	- Read the referenced sections in `design.md`.
105	- Confirm the implementation uses the prescribed structures, interfaces, and dependency direction.
106	- Reject silent substitutions for design-mandated choices.
107
108	### 10.5 Boundary Audit
109	- Compare the implementation against the design's boundary commitments and out-of-boundary statements.
110	- Reject if downstream-specific behavior is pushed into an upstream boundary for convenience.
111	- Reject if the implementation creates new hidden dependencies, shared ownership, or undeclared coupling across adjacent boundaries.

People who liked this also viewed

6 picks

Type