Agent skill

Codebase Impact Analysis

by openshift-eng in openshift-eng/ai-helpers

Analyze a Go codebase to determine if it is impacted by a specific CVE using multiple verification methods and assign a risk level

Apache-2.0Auto-check passedSecurity

Install Codebase Impact Analysis

skills CLI
$ npx skills add openshift-eng/ai-helpers --skill codebase-impact-analysis -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install openshift-eng/ai-helpers codebase-impact-analysis --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/openshift-eng/ai-helpers.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/compliance/skills/codebase-impact-analysis .claude/skills/codebase-impact-analysis && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
codebase-impact-analysis
GitHub stars
120
Token cost
~4.2k tokens
SKILL.md length
1,528 words
Files
1
Skills in repo
118
Repo updated
First seen
Licence
Apache-2.0

At a glance

Analyze a Go codebase to determine if it is impacted by a specific CVE using multiple verification methods and assign a risk level

  • Works in 4 steps: Identify Go Module Dependencies → Cross-Reference Vulnerable Packages → Build Evidence Package → …
  • Tasks that involve Vulnerability scanning
  • SKILL.md covers When to Use This Skill, Prerequisites, Implementation Steps and Return Value, plus 2 more sections
  • Calls go

What it does

Codebase Impact Analysis is an agent skill from openshift-eng/ai-helpers. Analyze a Go codebase to determine if it is impacted by a specific CVE using multiple verification methods and assign a risk level

Its SKILL.md is about 4.2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Security, covering Vulnerability scanning. It works with Go. The repository describes itself as: Developer productivity tools for Claude Code & other AI assistants. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve Vulnerability scanning

Example prompts

  • “/codebase-impact-analysis”

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. Identify Go Module Dependencies
  2. Cross-Reference Vulnerable Packages
  3. Build Evidence Package
  4. Assign Risk Level

What it can do on your machine

Read from SKILL.md and the folder at commit a627176. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • go

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Codebase Impact Analysis loads about 4.2k tokens when it runs. Until then it costs about 39 tokens; SKILL.md has 1,528 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~39
When it runs · the whole SKILL.md, loaded when a task matches
~4.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from openshift-eng/ai-helpers at commit a627176, republished under its Apache-2.0 licence (© openshift-eng). 1,528 words, ~4,224 tokens.

Download SKILL.mdSave it as .claude/skills/codebase-impact-analysis/SKILL.md (or your agent's skills folder).
name
codebase-impact-analysis
description
Analyze a Go codebase to determine if it is impacted by a specific CVE using multiple verification methods and assign a risk level

Codebase Impact Analysis

Determines whether a Go codebase is impacted by a specific CVE by applying multiple analysis methods with increasing confidence, collecting evidence, and assigning a risk level.

When to Use This Skill

Use this skill when:

  • A CVE profile has been gathered (from the cve-intelligence-gathering skill)
  • You need to determine if the current Go project is affected
  • You need to assign a risk level with supporting evidence

Prerequisites

Required Tools (validated in Phase 0 of the analyze-cve skill)
  • go toolchain with go.mod in workspace root
  • govulncheck: go install golang.org/x/vuln/cmd/govulncheck@latest
  • callgraph: go install golang.org/x/tools/cmd/callgraph@latest
  • digraph: go install golang.org/x/tools/cmd/digraph@latest
Required Inputs

From Phase 1 (cve-intelligence-gathering skill):

  • CVE ID
  • Affected package/module name(s)
  • Vulnerable version range
  • Fixed version (if available)
  • Vulnerable function signatures (if known)

From Parent Command:

  • --algo preference for call graph analysis (default: vta)
  • REPO_DIR — the repository cloned in Phase 0.7 (e.g. .work/compliance/analyze-cve/repos/hypershift). All commands below run against this directory, not necessarily the shell's current working directory.

Implementation Steps

Step 1: Identify Go Module Dependencies
bash
# Parse dependencies from go.mod
go list -m all

# Get detailed dependency info
go list -m -json all
  • Read go.mod from workspace root
  • Parse direct and indirect dependencies
  • Extract module versions
Step 2: Cross-Reference Vulnerable Packages

Apply the following methods in order. Each provides increasing confidence.

Method 1: Dependency Matching
  • Compare CVE-affected packages with go.mod dependencies
  • Check if affected package versions are in use
  • Account for version ranges and semantic versioning
bash
# Check if vulnerable package is a dependency
go list -m <vulnerable-package>

Decision Point:

  • IF package NOT in dependencies → Skip to risk assignment (likely LOW RISK)
  • IF package found → Continue to Method 2
Method 2: Go Vulnerability Scanner

CRITICAL RULES — read before running anything:

  1. Run govulncheck AT MOST ONCE per analysis run. Keep all Method 2 scratch and cache files under ${OUT_DIR}/ (same per-CVE workspace as call-graph artifacts). The canonical result is ${OUT_DIR}/govulncheck-source.txt — if it already exists and is non-empty for this run, read it and do not re-run.
  2. Never pipe govulncheck to head, tail, grep, or any other command. Always redirect to a file (> file 2>&1). Piping causes govulncheck to hang (SIGPIPE) when the reader closes.
  3. "No findings" is a valid and final result — it means the CVE is not yet in the Go vuln database. Proceed to Method 3 immediately. Do NOT re-run in a different mode or format.
  4. Always use timeout -k 10 to force-kill if SIGTERM is ignored. Plain timeout sends SIGTERM but govulncheck can ignore it when stuck in package loading.

This method has 4 sequential steps. If any step fails or times out, skip the remaining steps and proceed to Method 3 — govulncheck is one signal, not the only one.

bash
OUT_DIR="${OUT_DIR:-${AI_HELPERS_WORKSPACE:-.}/.work/compliance/analyze-cve/${CVE_ID}}"
mkdir -p "${OUT_DIR}"

Step 2a — go.mod check (instant)

bash
VULN_PKG="google.golang.org/grpc"   # replace with actual vulnerable package
echo "=== Step 2a: go.mod check for ${VULN_PKG} ==="
grep "${VULN_PKG}" "${REPO_DIR}/go.mod" && echo "FOUND in go.mod" || echo "NOT FOUND in go.mod"
  • IF NOT FOUND → record "package not in module graph" as LOW signal; skip Steps 2b–2d entirely; proceed to Method 3
  • IF FOUND → note the version; continue

Step 2b — Pre-flight: download modules and verify toolchain (max 2 min)

Large repos (300+ deps like spiffe-spire) need all modules cached before govulncheck can load them. Separate this from the scan to isolate network issues from analysis hangs.

bash
cd "${REPO_DIR}"
echo "=== Step 2b: Pre-flight ==="

# Download all modules (network-bound, do first)
echo "Downloading modules..."
timeout -k 10 120 env CGO_ENABLED=0 go mod download > "${OUT_DIR}/go-mod-download.txt" 2>&1
if [ $? -ne 0 ]; then
  echo "⚠ go mod download failed or timed out — govulncheck may fail"
  cat "${OUT_DIR}/go-mod-download.txt"
fi

# Verify the Go toolchain can load the package graph (CGO disabled first — many repos fail only with CGO enabled)
echo "Loading package list (CGO_ENABLED=0)..."
timeout -k 10 60 env CGO_ENABLED=0 go list ./... > "${OUT_DIR}/go-list-packages.txt" 2>&1
LIST_EXIT=$?
PKG_COUNT=$(wc -l < "${OUT_DIR}/go-list-packages.txt" 2>/dev/null || echo 0)
echo "Package count: ${PKG_COUNT}, exit code: ${LIST_EXIT}"

if [ $LIST_EXIT -ne 0 ]; then
  echo "go list with CGO_ENABLED=0 failed — retrying after CGO probe (Step 2c) before skipping govulncheck"
fi
  • IF go list succeeds with CGO_ENABLED=0 → continue to Step 2c, then 2d
  • IF go list still fails after Step 2c's CGO probe (with the chosen CGO_SETTING) → write the error to ${OUT_DIR}/govulncheck-source.txt, skip Steps 2c–2d, proceed to Method 3

Step 2c — CGO probe (max 60s, skip if compiler absent)

bash
echo "=== Step 2c: CGO probe ==="
CGO_SETTING=0
if command -v gcc >/dev/null 2>&1 || command -v cc >/dev/null 2>&1; then
  timeout -k 10 60 env CGO_ENABLED=1 go build ./... > "${OUT_DIR}/cgo-probe.txt" 2>&1
  if [ $? -eq 0 ]; then
    CGO_SETTING=1
    echo "✓ CGO works — using CGO_ENABLED=1"
  else
    echo "✗ CGO build failed — using CGO_ENABLED=0"
  fi
else
  echo "✗ No C compiler — using CGO_ENABLED=0"
fi
echo "CGO_ENABLED=${CGO_SETTING}"

if [ $LIST_EXIT -ne 0 ]; then
  echo "Retrying go list with CGO_ENABLED=${CGO_SETTING}..."
  timeout -k 10 60 env CGO_ENABLED=${CGO_SETTING} go list ./... > "${OUT_DIR}/go-list-packages.txt" 2>&1
  LIST_EXIT=$?
  PKG_COUNT=$(wc -l < "${OUT_DIR}/go-list-packages.txt" 2>/dev/null || echo 0)
  echo "Retry package count: ${PKG_COUNT}, exit code: ${LIST_EXIT}"
  if [ $LIST_EXIT -ne 0 ]; then
    echo "✗ go list failed after CGO probe — skipping govulncheck entirely"
    echo "go list failed (exit ${LIST_EXIT})" > "${OUT_DIR}/govulncheck-source.txt"
    cat "${OUT_DIR}/go-list-packages.txt" >> "${OUT_DIR}/govulncheck-source.txt"
  fi
fi
  • IF go list still fails after retry → ${OUT_DIR}/govulncheck-source.txt is populated; skip Step 2d and proceed to Method 3
  • IF go list succeeds → continue

Step 2d — govulncheck scan (max 5 min)

Use -scan=package first (fast, checks if CVE is in vuln DB and package imported). Only escalate to symbol-level if package-level finds something.

bash
if [ ! -s "${OUT_DIR}/govulncheck-source.txt" ] && [ $LIST_EXIT -eq 0 ]; then
  # Package-level scan first (fast — no symbol resolution)
  echo "=== Step 2d: govulncheck package scan ==="
  timeout -k 10 120 env CGO_ENABLED=${CGO_SETTING} govulncheck -scan=package ./... > "${OUT_DIR}/govulncheck-package.txt" 2>&1
  PKG_EXIT=$?
  echo "govulncheck -scan=package exit: ${PKG_EXIT}"
  cat "${OUT_DIR}/govulncheck-package.txt"

  # Check if the package scan found anything worth escalating to symbol level
  if grep -qi "Vulnerability\|finding\|${VULN_PKG}" "${OUT_DIR}/govulncheck-package.txt" 2>/dev/null; then
    echo "=== Step 2d: govulncheck symbol scan (escalating — CVE found at package level) ==="
    timeout -k 10 300 env CGO_ENABLED=${CGO_SETTING} govulncheck ./... > "${OUT_DIR}/govulncheck-source.txt" 2>&1
    SOURCE_EXIT=$?
    if [ $SOURCE_EXIT -eq 124 ] || [ $SOURCE_EXIT -eq 137 ]; then
      echo "govulncheck symbol scan timed out or was killed — using package-level results"
      cp "${OUT_DIR}/govulncheck-package.txt" "${OUT_DIR}/govulncheck-source.txt"
    fi
  else
    echo "Package scan found no findings — CVE likely not in Go vuln DB yet"
    cp "${OUT_DIR}/govulncheck-package.txt" "${OUT_DIR}/govulncheck-source.txt"
  fi
  echo "govulncheck complete"
else
  echo "=== govulncheck (using cached result) ==="
fi
cat "${OUT_DIR}/govulncheck-source.txt"

# Verify repo is still accessible after govulncheck
echo "=== Post-govulncheck repo check ==="
ls "${REPO_DIR}/go.mod" > /dev/null 2>&1 && echo "✓ Repo intact at ${REPO_DIR}" || echo "✗ WARNING: Repo missing at ${REPO_DIR}"
  • IF CGO was disabled → note in report: "CGO-gated code paths excluded from analysis"
  • IF package scan found no findings → CVE is not in Go vuln DB; do NOT escalate to symbol scan; proceed to Method 3
  • IF symbol scan timed out → use package-level results instead; proceed to Method 3
  • Save ${OUT_DIR}/govulncheck-source.txt as a workflow artifact

Decision Point — govulncheck is ONE signal. Always continue to Method 3 next.

  • IF scan reports vulnerable symbols called → Strong evidence for HIGH RISK; still continue to Method 3
  • IF scan reports no findings → CVE likely not yet in Go vuln database. Do NOT re-run. Proceed to Method 3.
  • IF scan timed out or was skipped → Proceed to Method 3; note the gap in the report
Method 3: Direct Dependency Check
bash
# Verify package is included (directly or transitively)
go list -mod=mod <vulnerable-package>

Note: Package presence alone doesn't prove vulnerable functions are called.

Method 4: Source Code Analysis
  • Search for import statements of vulnerable packages in source code
  • Use grep/codebase_search to find package usage
  • Search for vulnerable function/method names in codebase
  • Identify actual code paths that use vulnerable functions
  • Check if vulnerable functions are called in reachable code
Method 5: Call Graph Reachability Analysis (Mandatory when package is present)

Delegate to the call-graph-analysis skill.

  • Pass: --algo preference from user, vulnerable function signature, package path
  • Receive: Risk level, call chain, evidence files

Scope rule: Never invoke callgraph with ./.... Always target a specific main package (e.g. ./cmd/controller, .). The tool resolves transitive dependencies automatically. Running on ./... causes VTA to exhaust resources on repos with >50 packages (external-secrets has 138, spiffe-spire has 300+). See the call-graph-analysis skill for the progressive fallback chain (vta → rta → cha).

This method is REQUIRED whenever the vulnerable package is present in go.mod — regardless of what Methods 2, 3, or 4 found. Source code analysis (Method 4) is heuristic: it can miss indirect calls through interfaces, generated code, and runtime dispatch. Only a call graph provides provable reachability.

Valid reasons to skip call graph:

  • Package is NOT in go.mod (genuinely unreachable — LOW RISK by definition)
  • Codebase does not compile (note the limitation; rely on other methods)
  • Vulnerable function signature is unknown (note the gap; rely on govulncheck and source analysis)

NOT a valid reason to skip:

  • Source code analysis found no direct calls to the vulnerable function
  • govulncheck did not flag it (CVE may not be in the Go vuln DB yet)
  • The analysis "feels" complete from earlier methods
Show full SKILL.md (552 more words)Show less
Method 6: Configuration and Context Analysis
  • Review if vulnerable features are actually enabled
  • Check if vulnerable code paths are behind feature flags
  • Verify if inputs can reach vulnerable functions
  • Consider security controls (input validation, sandboxing)
Confidence Levels

Each method provides increasing confidence:

  1. Basic Presence (Low) — Package in go.mod (Method 1, 3)
  2. Import & Version Analysis (Medium) — Package imported, version in vulnerable range, function names found (Method 4)
  3. Vulnerability Scanner (Medium-High) — govulncheck confirms reachable vulnerable symbols (Method 2)
  4. Call Graph Reachability (Definitive) — Proven execution path from entry point to vulnerable function (Method 5)
  5. Context Analysis — Mitigating or aggravating factors (Method 6)

Required minimum: If the package is in go.mod, the analysis is not complete until Method 5 has run or a valid skip reason has been documented. Never stop at Method 4 alone.

Step 3: Build Evidence Package

Collect evidence from all methods used:

  • Dependency Evidence: go.mod entries, go list output, version info
  • Static Code Evidence: File paths, line numbers, code snippets showing usage
  • Reachability Evidence: Call graph output, execution paths, DOT visualization (saved to ${AI_HELPERS_WORKSPACE:-.}/.work/compliance/analyze-cve/{CVE-ID}/callgraph.svg, outside REPO_DIR)
  • Scanner Evidence: govulncheck output, vulnerability findings
  • Mitigation Factors: Input validation, disabled features, feature flags, security controls
Step 4: Assign Risk Level

Evaluate all evidence and assign a risk level. The determination should be data-driven, not formula-based.

HIGH RISK:

  • Call graph shows a reachable path to the vulnerable function, OR
  • Symbol-level govulncheck confirms called vulnerable symbols (package-level import alone is not sufficient)

MEDIUM RISK:

  • Package + vulnerable version in dependencies, usage evidence present, but call graph could not run (build failure or unknown function signature) — reachability not definitively proven

LOW RISK:

  • Package not in dependencies (call graph skipped — genuinely unreachable), OR version not in vulnerable range, OR call graph ran and found no reachable path

NEEDS REVIEW:

  • Package is in go.mod but call graph was skipped for any reason other than package absence or build failure — escalate; do not leave as LOW based on source code analysis alone
  • Conflicting signals or incomplete analysis

Rule: If package is in go.mod and call graph was skipped because source analysis "found nothing", assign NEEDS REVIEW, not LOW RISK. Document the skip reason explicitly.

Return Value

Return structured result to parent command:

json
{
  "skill": "codebase-impact-analysis",
  "status": "success",
  "risk_level": "<HIGH|MEDIUM|LOW|NEEDS_REVIEW>",
  "methods_used": ["dependency_matching", "govulncheck", "direct_dependency_check", "source_code_analysis", "call_graph", "context_analysis"],
  "evidence": {
    "dependency": {
      "package_found": true,
      "current_version": "<version>",
      "dependency_type": "<direct|indirect>",
      "in_vulnerable_range": true
    },
    "govulncheck": {
      "ran": true,
      "cve_found": true,
      "vulnerable_symbols_called": true
    },
    "call_graph": {
      "ran": true,
      "algorithm": "<vta|rta|cha|static>",
      "reachable_from_main": true,
      "call_chain": "main -> handler -> parse -> VULN",
      "evidence_files": ["callgraph.dot", "callgraph.svg"],
      "skip_reason": "<null if ran | 'package_not_in_gomod' | 'build_failure' | 'unknown_function_signature'>"
    },
    "source_analysis": {
      "import_found": true,
      "function_usage_found": true,
      "files": ["<file1>:<line>", "<file2>:<line>"]
    },
    "mitigation_factors": []
  },
  "confidence_assessment": {
    "level": "<HIGH|MEDIUM|LOW>",
    "methods_count": 4,
    "gaps": ["<any gaps in analysis>"]
  }
}

Error Handling

Build Failures
  • IF project doesn't compile → Note limitation, skip call graph analysis, rely on other methods
Missing CVE in govulncheck Database
  • IF govulncheck doesn't know about this CVE → Continue with other methods, note gap
Large Codebases
  • IF call graph times out → Follow fallback strategy in call-graph-analysis skill: algorithm fallback (vta → rta → cha), always targeting a specific main package. Never use ./... as scope.
Incomplete CVE Information
  • IF vulnerable function signature unknown → Skip call graph, note skip_reason: unknown_function_signature, assign MEDIUM at best — do NOT assign LOW based on source analysis alone
Source Code Analysis Shows No Usage
  • This is NOT a valid reason to skip call graph. Proceed with Method 5.
  • Source code search misses interface dispatch, generated code, and indirect call paths.

Integration with analyze-cve

This skill is called from Phase 2 of the analyze-cve skill, after the Repo Guard has confirmed REPO_DIR still exists.

Input: CVE profile from Phase 1, --algo preference from user, REPO_DIR set in Phase 0.7 Output: Risk level, evidence package, confidence assessment Next: The analyze-cve skill uses risk level to decide whether to generate report and proceed to remediation

© openshift-eng, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in plugins/compliance/skills/codebase-impact-analysis of openshift-eng/ai-helpers.

Open the folder on GitHubat commit a627176

Compare with similar skills

Codebase Impact Analysis next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Codebase Impact Analysis compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Codebase Impact Analysis this skillopenshift-eng/ai-helpers120—~4.2kAutomated safety check: PassApache-2.0
Snapshotboostsecurityio/poutine522—~214Automated safety check: PassApache-2.0
Dependency Vulnerability AuditHabitat-Thinking/ai-literacy-superpowers114—~2.1kAutomated safety check: PassCustom licence
Golang Pkg Go Devcontext-labs/whip1.1k2 repos~3kAutomated safety check: PassMIT
Docsboostsecurityio/poutine522—~336Automated safety check: PassApache-2.0
Cve Remediator V2kubernetes-sigs/cloud-provider-azure294—~2.5kAutomated safety check: PassApache-2.0

Similar skills

  • Snapshot

    boostsecurityio/poutine

    Run snapshot regression tests after changes to OPA rules, scanners, analyzers, or formatters to detect output regressions.

    522 GitHub stars~214 tokensUpdated yesterday
    SecurityAuto-check passed
  • Dependency Vulnerability Audit

    Habitat-Thinking/ai-literacy-superpowers

    A skill your agent uses when auditing project dependencies for known vulnerabilities, supply chain risk, or provenance issues — covers Go modules, Maven/JVM, and CI integration for automated scanning

    114 GitHub stars~2.1k tokensUpdated 17 days ago
    SecurityAuto-check passed
  • Golang Pkg Go Dev

    context-labs/whip

    Golang package/module docs via godig, a pkg.go.dev API client (CLI + MCP) — APIs, symbols, versions, importers, licenses, vulnerabilities.

    1.1k GitHub starsUsed in 2 repos~3k tokens
    SecurityAuto-check passed
  • Docs

    boostsecurityio/poutine

    Update project documentation when features are added or changed.

    522 GitHub stars~336 tokensUpdated yesterday
    SecurityAuto-check passed
  • Cve Remediator V2

    kubernetes-sigs/cloud-provider-azure

    Official

    Raise Go modules to caller-supplied minimum fixed versions from CVE/GO findings in any format, per tracked module root, sync go.mod/go.sum and root vendor/, audit the source module graphs, run the…

    294 GitHub stars~2.5k tokensUpdated today
    SecurityAuto-check passed
  • Update Vulndb

    boostsecurityio/poutine

    Update the embedded build platform vulnerability database from the CVE Project's cvelistV5 repository.

    522 GitHub stars~173 tokensUpdated yesterday
    SecurityAuto-check passed

More from openshift-eng/ai-helpers

All 118 skills in this repo
  • Investigate CI Reliability

    openshift-eng/ai-helpers

    Find and independently validate actionable reliability defects across OpenShift release jobs and presubmits, then export portable issue handoffs.

    120 GitHub stars~1.9k tokensUpdated yesterday
    Auto-check passed
  • Address Review PR

    openshift-eng/ai-helpers

    Fetch and address all PR review comments — categorize by priority, make code changes, post replies, and push.

    120 GitHub stars~2.9k tokensUpdated yesterday
    Auto-check passed
  • Categorize Activity Types

    openshift-eng/ai-helpers

    Categorize Jira issues into Red Hat Sankey Activity Type categories using MCP Jira tools.

    120 GitHub stars~2.4k tokensUpdated yesterday
    Auto-check passed
  • Has Review Work

    openshift-eng/ai-helpers

    Decide whether a GitHub PR has unanswered authorized review comments or new required CI failures worth a follow-up agent.

    120 GitHub stars~1.9k tokensUpdated yesterday
    Auto-check passed
  • Must Gather Analyzer

    openshift-eng/ai-helpers

    Analyze OpenShift must-gather diagnostic data including cluster operators, pods, nodes, and network components.

    120 GitHub stars~2.3k tokensUpdated yesterday
    Auto-check passed
  • Payload Autodl JSON

    openshift-eng/ai-helpers

    Schema for the autodl JSON data file produced by payload-analysis for database ingestion — you must use this skill whenever generating the autodl JSON file

    120 GitHub stars~2.6k tokensUpdated yesterday
    Auto-check passed

Works with

Categories

Questions about Codebase Impact Analysis

What does Codebase Impact Analysis do?

Analyze a Go codebase to determine if it is impacted by a specific CVE using multiple verification methods and assign a risk level. Codebase Impact Analysis is an agent skill from openshift-eng/ai-helpers.

When should I use Codebase Impact Analysis?

Codebase Impact Analysis fits situations like: tasks that involve Vulnerability scanning.

How do I install Codebase Impact Analysis in Claude Code?

Run `npx skills add openshift-eng/ai-helpers --skill codebase-impact-analysis -a claude-code`. Or copy the skill folder (plugins/compliance/skills/codebase-impact-analysis in openshift-eng/ai-helpers) into .claude/skills/codebase-impact-analysis in your project. Claude Code loads it when a task matches its description.

How do I install Codebase Impact Analysis in Codex?

Run `npx skills add openshift-eng/ai-helpers --skill codebase-impact-analysis -a codex`. Or copy the skill folder (plugins/compliance/skills/codebase-impact-analysis in openshift-eng/ai-helpers) into .agents/skills/codebase-impact-analysis in your project. Codex loads it when a task matches its description.

Can I use Codebase Impact Analysis in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add openshift-eng/ai-helpers --skill codebase-impact-analysis -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/codebase-impact-analysis, .gemini/skills/codebase-impact-analysis, .github/skills/codebase-impact-analysis and .opencode/skills/codebase-impact-analysis in your project.

What does Codebase Impact Analysis need to run?

Going by SKILL.md and its folder, Codebase Impact Analysis needs the command-line tools its instructions call (go).

Does Codebase Impact Analysis access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Codebase Impact Analysis safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Codebase Impact Analysis use?

Codebase Impact Analysis is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Codebase Impact Analysis use?

About 4.2k tokens (SKILL.md is roughly 17k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Codebase Impact Analysis?

Skills that share tags, products or a category with Codebase Impact Analysis: Snapshot (boostsecurityio/poutine, 522 stars), Dependency Vulnerability Audit (Habitat-Thinking/ai-literacy-superpowers, 114 stars), Golang Pkg Go Dev (context-labs/whip, 1.1k stars) and Docs (boostsecurityio/poutine, 522 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Codebase Impact Analysis?

openshift-eng (a GitHub organization) maintains it in openshift-eng/ai-helpers, which has 120 GitHub stars. The repository holds 118 skills in this directory. The repository was last updated on October 6, 2026.

Source: openshift-eng/ai-helpers on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.