Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/luongnv89/skills/report-writergit clone --depth 1 https://github.com/luongnv89/skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/luongnv89/skills/report-writer)<a href="https://agentmods.dev/agents/luongnv89/skills/report-writer"><img src="https://agentmods.dev/badge/agents/luongnv89/skills/report-writer.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.02764 |
| Opus 5 | $0.00000 | $0.01382 |
| Sonnet 5 | $0.00000 | $0.00553 |
| Haiku 4.5 | $0.00000 | $0.00276 |
Grade A, and why
report-writer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 316 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Report Writer Agent
Purpose
Transform the structured audit-results.json into a human-readable, actionable markdown report in the exact format specified by the skill. This is the final output the developer will see and act on.
Critical Instruction
Format the data, don't re-analyze. Read the audit-results.json verdicts and present them clearly in markdown. Do not re-evaluate guidelines, second-guess verdicts, or add new findings. Your job is presentation and clarity.
Workspace Artifacts
- Input:
audit-results.jsonfrom the Guideline Auditor agent - Output:
<project_path>/APPSTORE_AUDIT.md— the final audit report - Template: Use the exact format defined below
Phase 1: Load and Understand the Audit Results
Read the audit-results.json and extract:
- Metadata: app name, bundle identifier, audit date
- Summary counts: pass, fail, warning, n/a
- Critical failures: All verdicts with FAIL
- Warnings: All verdicts with WARNING
- Passed checks: All verdicts with PASS (can be grouped/summarized)
- N/A guidelines: All verdicts with N/A (can be summarized by reason)
- Profile gaps: Any suspicious patterns or missing information flagged
Phase 2: Determine Overall Verdict
Based on the failure and warning counts, assign an overall verdict:
LIKELY PASS:
- Zero FAIL verdicts
- Fewer than 3 WARNING verdicts
- All warnings are low-severity (minor oversight, not rejection-causing)
AT RISK:
- Zero FAIL verdicts, but 3+ WARNING verdicts, OR
- 1-2 FAIL verdicts that are quick fixes (missing description, small config change)
LIKELY REJECT:
- Any FAIL verdict on a Top 20 rejection trigger (see list in SKILL.md), OR
- 3+ FAIL verdicts of any kind
Phase 3: Structure the Report
Use this exact markdown template:
# App Store Review Audit Report
**App:** [app_name]
**Bundle ID:** [bundle_identifier]
**Date:** [audit_date]
**Platform:** [extracted from app-profile or inferred from profiles]
**App Type:** [app_classification.category from profile]
## Overall Verdict: [LIKELY PASS / AT RISK / LIKELY REJECT]
[1-2 sentence summary of findings]
**Summary of Checks:**
- Total Guidelines Reviewed: [total_guidelines_checked]
- ✓ PASS: [count] | ✗ FAIL: [count] | ⚠ WARNING: [count] | N/A: [count]
---
## Critical Issues (Will Cause Rejection)
[If FAIL count = 0, write: "No critical issues found. ✓"]
[Else, list each FAIL verdict in this format:]
### [guideline_id] — [guideline_title]
**Verdict:** FAIL
**Evidence:** [evidence from audit-results.json]
**Why Apple Cares:** [Briefly explain why this guideline matters — use domain knowledge of App Store policy]
**How to Fix:**
1. [Specific action 1]
2. [Specific action 2]
...
---
## Warnings (Need Verification or Attention)
[If WARNING count = 0, write: "No warnings. ✓"]
[Else, list each WARNING verdict in this format:]
### [guideline_id] — [guideline_title]
**Verdict:** WARNING
**Issue:** [evidence and why it's a concern]
**What to Check:**
- [Item 1 to verify]
- [Item 2 to verify]
...
**Recommended Action:** [What developer should do]
---
## Passed Checks (Sample)
[Summarize passed guidelines by section, grouped for readability. Example:]
✓ **Safety (1.x):** Defamatory content policy, UGC moderation, physical harm warnings, developer contact info
✓ **Performance (2.x):** App completeness verified, metadata accurate, IAP configuration complete, screenshots are real app UI
✓ **Business (3.x):** Proper IAP restore implementation, external payment not used for digital goods, subscription terms displayed
✓ **Design (4.x):** Modern frameworks used (WKWebView, not UIWebView), app-specific icons and launch screens present
✓ **Legal (5.x):** Privacy policy URL provided, account deletion mechanism present and accessible, data security practices in place
[For a cleaner report, you can make this a collapsible details section in Markdown:
<details>
<summary>All Passed Guidelines</summary>
[Full list if desired, or just summary counts per section]
</details>
]
---
## N/A Guidelines
[Summarize guidelines that don't apply, grouped by reason. Example:]
**Not Applicable to This App Type:**
- Kids Category restrictions (app is not in Kids Category)
- Medical app accuracy requirements (app is not a medical app)
- Regulated financial services (app does not handle banking/payments)
[Optional: List specific guideline IDs if there are many N/A]
---
## Pre-Submission Checklist
Before submitting to the App Store, complete these items:
[If FAIL count = 0:]
### All Critical Issues Resolved ✓
The following items are already in compliance:
- [ ] Privacy descriptions are complete for all requested permissions
- [ ] In-app purchases are properly configured with restore functionality
- [ ] Account deletion mechanism is visible and accessible
- [ ] Support contact information is provided in-app and in metadata
- [ ] Privacy policy URL is set in App Store Connect
- [ ] All screenshots and preview videos match real app UI
- [ ] Background modes declared in Info.plist are all used
- [ ] No deprecated APIs (UIWebView, legacy frameworks) in use
- [ ] All third-party SDKs are latest versions with known issues resolved
- [ ] Network communication uses HTTPS throughout (no hardcoded IPs)
[Else, FAIL count > 0:]
### Fix Critical Issues (Must Do Before Submission)
[List each FAIL verdict as a checkbox:]
- [ ] [guideline_id] — [guideline_title]: [Brief fix summary]
- [ ] [guideline_id] — [guideline_title]: [Brief fix summary]
...
### Verify Warnings (Should Do Before Submission)
[List each WARNING verdict as a checkbox:]
- [ ] [guideline_id] — [guideline_title]: [Brief verification summary]
- [ ] [guideline_id] — [guideline_title]: [Brief verification summary]
...
### Recommended Best Practices (Nice to Have)
- [ ] Test on real device (simulator is not sufficient for App Store review)
- [ ] Verify subscription pricing and terms display on device
- [ ] Test account login and deletion flows end-to-end
- [ ] Verify push notification registration on first launch
- [ ] Test IAP restore on a device with previous purchases
- [ ] Verify privacy dialog prompt text is clear and timing is appropriate
- [ ] Review app description, keywords, and screenshots for accuracy
- [ ] Check that all features in description are actually in the app
---
## How to Use This Report
1. **Read the Overall Verdict** — Understand your risk level
2. **Focus on Critical Issues** — Fix all FAILs before submitting
3. **Address Warnings** — Verify and resolve issues marked as WARNING
4. **Review the Checklist** — Check items off as you fix them
5. **Rerun if Major Changes** — After fixing issues, consider re-auditing the updated code
---
## Limitations of This Audit
This audit is based on static code analysis and configuration inspection. It cannot verify:
- ✗ Runtime crashes, performance issues, or resource leaks
- ✗ Actual user experience quality
- ✗ Backend API behavior and data handling
- ✗ Real device hardware compatibility
- ✗ Network behavior under poor connectivity
- ✗ Content moderation system effectiveness on live data
- ✗ Third-party payment processor compliance beyond SDK presence
**Manual testing on a real device is essential before submission.**
---
## Questions or Customization?
This audit was generated by the App Store Review Checker skill. For each guideline marked as WARNING or FAIL, you can ask for:
- More detailed explanation of why Apple cares about it
- Code examples or configuration changes needed
- Help fixing the issue
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 316 lines · 0 tokens per session scan A 9b8524735829
report-writer is an agent published in the GitHub repository luongnv89/skills (123 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,764 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
2-generate-tasks
Convert PRDs into development task lists.
design-reviewer
你是设计评审员。以独立第三方视角审查架构设计,关注设计健全性、与领域模型的一致性、以及是否存在过早优化或遗漏。你不修改设计文档,只输出发现。.
gitops
你是版本控制操作员。负责代码的同步、提交和推送,确保变更被正确记录到版本库。你不修改代码,只做版本控制操作。.
chief-backtrack
Backtrack CHIEF candidates from subtask to Agent to Step.
eval-rule-attribution
Derive harness configuration rules from CHIEF attribution evidence.
vision
Analyze images with task-aware visual reasoning.