Skills ghe-requirements
📦
ghe-requirements
Content revision r1 High Risk ⚙️ External commands📁 Filesystem access
Manage GHE Requirements
Feature work can drift when requirements are missing or stale. This skill defines requirement files, versioning, issue links, and backup steps.
Supports: Claude Codex Code(CC)
Stop for confirmation before installing.
Review the plan and obtain explicit user consent before changing files.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "ghe-requirements" from https://skillstore.io/skills/emasoft-ghe-requirements.md and its manifest at https://skillstore.io/api/skills/emasoft-ghe-requirements/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "ghe-requirements". Create a requirement for adding export filters to a reporting dashboard.
Expected outcome:
- A draft requirement with a feature overview, user story, and acceptance criteria for filtered exports.
- A proposed atomic change list covering UI controls, export behavior, and tests.
- A dependency section noting related dashboard and reporting requirements.
Using "ghe-requirements". Check whether issue 42 is linked to the current REQ version.
Expected outcome:
- A validation summary showing whether the issue includes a requirement link.
- A version comparison between the issue text and the current requirement file.
- A short remediation note if the issue is missing or stale.
Security Audit
High Riskv9 • 7/5/2026 Open versioned report
The audit found many Markdown backtick false positives, but several bash workflow blocks remain real external-command surfaces. The highest confirmed risk is authorization to write ~/.claude settings, plus an overbroad instruction block that can weaken normal instruction hierarchy.
1
Files scanned
327
Lines analyzed
29
Review items
0
False positives ignored
Confirmed security concerns (1)
High
Prompt Injection Attempt Detected
SKILL.md declares an absolute "IRON LAW" that every user word must be followed with no exceptions. This attempts to override normal instruction hierarchy and can pressure an agent to obey unsafe user demands.
The text uses absolute authority language and tells the agent to follow every user statement without exceptions. It does not explicitly mention system prompts, so confidence is high but not maximal.
Capability review items (29)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
High
Hidden file in home directory
- ~/.claude (for plugin/settings fixes)
Line 26 explicitly permits writes to ~/.claude for plugin or settings fixes. Allowing a marketplace skill to modify hidden home-directory agent settings is a real persistence and configuration-tampering risk.
High
Hidden file access
- ~/.claude (for plugin/settings fixes)
Line 26 explicitly references ~/.claude, a hidden home-directory configuration path. Access to this path can affect agent behavior outside the project workspace.
Medium
Ruby/shell backtick execution
```bash
This bash block instructs agents to run shell commands that scan files and create a requirement file from variable-controlled path parts. That is legitimate workflow guidance, but it is a real command-execution surface.
Medium
Ruby/shell backtick execution
```bash
This bash block edits a requirement file with sed and commits the result with git. It can modify repository files and history, so it is a real command-execution surface.
Medium
Ruby/shell backtick execution
```bash
This bash block reads versions, edits files with sed, and commits changes with git. It can modify repository history and files, so the command guidance is security-relevant.
Medium
Ruby/shell backtick execution
```bash
This bash block creates a backup directory and copies requirement files into .serena memory. It performs filesystem writes based on globbed repository paths.
Medium
Ruby/shell backtick execution
```bash
This bash block invokes gh issue create with generated title, labels, and body content. It can create remote GitHub issues and publish requirement details.
Medium
Ruby/shell backtick execution
```bash
This bash block searches .git files and posts GitHub issue comments for each derived item. It can read repository internals and mutate remote issue discussions.
Medium
Ruby/shell backtick execution
```bash
This bash block invokes gh issue view and validates the returned body. It performs remote reads using a variable issue identifier.
Medium
Ruby/shell backtick execution
```bash
This bash block extracts a requirement path from issue text and checks local file existence. The local path is derived from external issue content.
Medium
Ruby/shell backtick execution
```bash
This bash block compares issue-linked and local requirement versions. It reads a file path derived from prior issue parsing, so it needs path validation.
Medium
Shell command substitution
LAST_REQ=$(find REQUIREMENTS -name "REQ-*.md" | sed 's/.*REQ-\([0-9]*\).*/\1/' | sort -n | tail -1)
The command substitution runs find, sed, sort, and tail across REQUIREMENTS to choose the next ID. It is a real shell pipeline over repository files, so it should be reviewed before use.
Medium
Shell command substitution
NEXT_REQ=$(printf "%03d" $((LAST_REQ + 1)))
The command substitution executes printf with arithmetic derived from the prior file scan. The command is narrow, but it still runs shell logic generated by the skill.
Medium
Shell command substitution
CURRENT_VERSION=$(grep "^version:" "$REQ_FILE" | cut -d' ' -f2)
The command substitution reads a version from a variable-controlled requirement file path. The path is quoted, but the workflow still performs shell file reads based on agent state.
Medium
Shell command substitution
NEW_VERSION=$(echo $CURRENT_VERSION | awk -F. '{print $1"."$2+1".0"}')
The command substitution runs awk on the current version string to create a new version. This appears legitimate, but it is shell processing that affects later file edits.
Medium
Shell command substitution
cp "$req" ".serena/memories/requirements/$(basename $req)"
The command substitution uses basename while copying requirement files into a backup directory. It is a real filesystem write loop over repository file names.
Medium
Shell command substitution
--body "$(cat <<EOF
The command substitution builds a GitHub issue body for gh issue create. It can publish generated content to GitHub, so it is a real external command surface.
Medium
Shell command substitution
LINKED_ISSUES=$(grep -l "REQ-${REQ_NUM}" .git/*) # Or use GitHub search
The command substitution searches .git files for requirement references and feeds results into issue updates. Reading repository internals and deriving issue targets from grep output is risky.
Medium
Shell command substitution
ISSUE_BODY=$(gh issue view $ISSUE --json body --jq '.body')
The command substitution calls gh issue view with a variable issue identifier. This is a legitimate validation step, but it performs a remote GitHub read from variable input.
Medium
Shell command substitution
REQ_PATH=$(echo "$ISSUE_BODY" | grep -oP 'REQUIREMENTS/[^\s\)]+\.md' | head -1)
The command substitution extracts a local requirement path from GitHub issue body text. Because issue text can be user-controlled, later filesystem checks depend on external content.
Medium
Shell command substitution
CURRENT_VERSION=$(grep "^version:" "$REQ_PATH" | cut -d' ' -f2)
The command substitution reads a requirement file path derived from issue content. That creates a local file-read surface controlled by the linked issue text.
Medium
Template literal with command substitution
```bash
This bash block instructs agents to run shell commands that scan files and create a requirement file from variable-controlled path parts. That is legitimate workflow guidance, but it is a real command-execution surface.
Medium
Template literal with command substitution
```bash
This bash block reads versions, edits files with sed, and commits changes with git. It can modify repository history and files, so the command guidance is security-relevant.
Medium
Template literal with command substitution
```bash
This bash block creates a backup directory and copies requirement files into .serena memory. It performs filesystem writes based on globbed repository paths.
Medium
Template literal with command substitution
```bash
This bash block invokes gh issue create with generated title, labels, and body content. It can create remote GitHub issues and publish requirement details.
Medium
Template literal with command substitution
```bash
This bash block searches .git files and posts GitHub issue comments for each derived item. It can read repository internals and mutate remote issue discussions.
Medium
Template literal with command substitution
```bash
This bash block invokes gh issue view and validates the returned body. It performs remote reads using a variable issue identifier.
Medium
Template literal with command substitution
```bash
This bash block extracts a requirement path from issue text and checks local file existence. The local path is derived from external issue content.
Medium
Template literal with command substitution
```bash
This bash block compares issue-linked and local requirement versions. It reads a file path derived from prior issue parsing, so it needs path validation.
Risk Factors
⚙️ External commands (49)
SKILL.md:39 SKILL.md:40 SKILL.md:54 SKILL.md:58-70 SKILL.md:70-76 SKILL.md:76-128 SKILL.md:128-134 SKILL.md:134-141 SKILL.md:141-153 SKILL.md:153-160 SKILL.md:160-166 SKILL.md:166-170 SKILL.md:170-182 SKILL.md:182-198 SKILL.md:198-204 SKILL.md:204-212 SKILL.md:212-223 SKILL.md:223-241 SKILL.md:241-247 SKILL.md:247-263 SKILL.md:263-269 SKILL.md:269-276 SKILL.md:276-280 SKILL.md:280-287 SKILL.md:287-291 SKILL.md:291-301 SKILL.md:301-321 SKILL.md:321-322 SKILL.md:322-326 SKILL.md:136 SKILL.md:137 SKILL.md:184 SKILL.md:187 SKILL.md:191 SKILL.md:209 SKILL.md:227-230 SKILL.md:249 SKILL.md:271 SKILL.md:282 SKILL.md:293 SKILL.md:296 SKILL.md:134-141 SKILL.md:182-198 SKILL.md:204-212 SKILL.md:223-241 SKILL.md:247-263 SKILL.md:269-276 SKILL.md:280-287 SKILL.md:291-301
📁 Filesystem access (4)
Audited by: codex View Audit History →
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/emasoft-ghe-requirements/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/emasoft-ghe-requirements?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/emasoft-ghe-requirements?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/emasoft-ghe-requirements/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/emasoft-ghe-requirements.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
Emasoft. (2026). ghe-requirements security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/emasoft-ghe-requirements/audits/9BibTeX citation
@techreport{emasoft-emasoft-ghe-requirements-2026,
author = {Emasoft},
title = {ghe-requirements security audit report (audit version 9)},
institution = {Skillstore},
year = {2026},
number = {9},
url = {https://skillstore.io/skills/emasoft-ghe-requirements/audits/9},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "ghe-requirements security audit report (audit version 9)"
version: "unspecified"
type: report
authors:
- name: "Emasoft"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/emasoft-ghe-requirements/audits/9"
identifiers:
- type: other
value: "skillstore:emasoft-ghe-requirements:audit:9"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: Medium 55
Architecture
85
Maintainability
87
Content
70
Community
83
Spec Compliance
What You Can Build
Prepare a Feature Requirement
Create a structured requirement before development starts, with acceptance criteria, atomic changes, tests, and dependencies.
Coordinate GitHub Issue Work
Link GitHub development issues to the current requirement version so implementers share the same source of truth.
Review Requirement Drift
Compare issue links, requirement versions, and acceptance criteria before implementation review or test planning.
Try These Prompts
Create a Requirement Draft
Create a GHE requirement draft for this feature request: [feature]. Include user story, acceptance criteria, atomic changes, tests, and dependencies.
Update an Existing Requirement
Update this GHE requirement for a minor version change: [requirement summary]. Preserve original intent and list the revision history entry.
Link a Requirement to an Issue
Prepare GitHub issue requirements text for REQ-[number] version [version]. Include the primary link, acceptance criteria, and atomic changes.
Audit Requirement Coverage
Audit this GitHub issue against its linked GHE requirement. Identify missing links, stale versions, uncovered acceptance criteria, and unclear tests.
Best Practices
- Confirm requirement changes with the user before changing acceptance criteria or scope.
- Keep issue links tied to explicit requirement versions.
- Review shell examples before running them in repositories with sensitive files.
Avoid
- Starting development work before a requirement file exists.
- Changing requirement scope without updating revision history.
- Running GitHub CLI or file commands without checking paths and permissions.
Frequently Asked Questions
What does this skill manage?
It manages GHE requirement documents, version history, issue links, validation checks, and backup guidance.
Does it create files automatically?
No. It provides guidance and examples that an agent or user may apply after review.
Which tools can use this skill?
The report lists support for Claude, Codex, and Claude Code.
Does it require GitHub CLI access?
Some examples use GitHub CLI commands, so those workflows require authentication and repository permissions.
Can it validate linked requirement versions?
Yes. It describes checks for requirement links, file existence, and version drift.
What should users review before use?
Users should review file paths, shell variables, GitHub issue targets, and repository permissions before running commands.