icarosama/Kenoma

Monitor repositories for security vulnerability patches using AI

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
api-keyAPI key for the model provider (OpenAI, DeepSeek, etc.)no
anthropic-api-keyAnthropic API key (deprecated: use api-key instead)no
providerModel provider: anthropic, openai, deepseek, or openai-compatible. Defaults to anthropic if omitted.noanthropic
base-urlCustom base URL for openai-compatible providers (e.g. http://localhost:11434/v1 for Ollama)no""
github-tokenGitHub token for API accessyes${{ github.token }}
repositoriesJSON array of repositories to monitor (e.g., [{"owner":"expressjs","repo":"express"}])yes
state-filePath to state file for tracking last checked commitsno.vulnerability-spoiler-state.json
create-issuesWhether to create GitHub issues for detected vulnerabilitiesnotrue
issue-repoRepository to create issues in (defaults to current repository)no""
modelModel to use for analysis (e.g. claude-sonnet-4-6, gpt-4o, deepseek-chat). Must match your provider.noclaude-sonnet-4-6
max-commitsMaximum number of commits to analyze per repository per runno50
enable-repo-contextFetch modified files for context (max 3 files, 3000 chars each)nofalse
enable-judgeReview detections with second model (only runs on positives)nofalse
judge-modelModel for judge (empty = same as primary model; uses the same provider)no""
max-concurrencyMaximum number of commits to analyze in parallel per repositoryno3
path-filterComma-separated path prefixes; only analyze commits that touch at least one matching file (empty = analyze all)no""
skip-authorsComma-separated author name substrings to skip (e.g. "[bot],dependabot")no""
llm-timeout-secondsSeconds before an LLM API call is aborted and retriedno60
namedescription
vulnerabilities-foundNumber of vulnerabilities detected
issues-createdNumber of issues created
analyzed-commitsTotal number of commits analyzed
resultsJSON array of detected vulnerabilities with details