Security scanner for AI agent skills & MCP servers. 173 detection rules. 13 categories. 5 registries monitored daily. OpenClaw detection included. No API keys, no cloud, no LLM. One binary. Detection engine behind Oktsec.

ai-agents claude claude-code golang mcp openclaw openclaw-skills prompt-injection security skills
5 Open Issues Need Help Last updated: Mar 5, 2026

Open Issues Need Help

View All on GitHub

AI Summary: This GitHub issue aims to enhance the user experience of the `aguara scan` command. When provided with a non-existent path, the tool currently displays a generic operating system error. The goal is to replace this with a clear, user-friendly message indicating that the specified path does not exist, along with appropriate test coverage.

Complexity: 2/5
bug good first issue

Security scanner for AI agent skills & MCP servers. 173 detection rules. 13 categories. 5 registries monitored daily. OpenClaw detection included. No API keys, no cloud, no LLM. One binary. Detection engine behind Oktsec.

Go
#ai-agents#claude#claude-code#golang#mcp#openclaw#openclaw-skills#prompt-injection#security#skills

AI Summary: The issue requests adding a `--json` boolean flag to the `aguara scan` command. This flag would act as a shorthand for `--format json`, following common CLI conventions, and should conflict if used alongside an explicit `--format` flag.

Complexity: 2/5
enhancement good first issue

Security scanner for AI agent skills & MCP servers. 173 detection rules. 13 categories. 5 registries monitored daily. OpenClaw detection included. No API keys, no cloud, no LLM. One binary. Detection engine behind Oktsec.

Go
#ai-agents#claude#claude-code#golang#mcp#openclaw#openclaw-skills#prompt-injection#security#skills

AI Summary: This issue proposes adding support for YAML front-matter in markdown files to allow per-file configuration of `aguara` behavior. Users would define settings like disabling specific rules within the front-matter block. The implementation requires modifying the parser to extract this configuration and updating the rule engine to skip disabled rules for the respective file, with comprehensive test coverage.

Complexity: 3/5
enhancement good first issue

Security scanner for AI agent skills & MCP servers. 173 detection rules. 13 categories. 5 registries monitored daily. OpenClaw detection included. No API keys, no cloud, no LLM. One binary. Detection engine behind Oktsec.

Go
#ai-agents#claude#claude-code#golang#mcp#openclaw#openclaw-skills#prompt-injection#security#skills

AI Summary: This issue requests the addition of an `aguara completion` subcommand to generate shell completion scripts. It should support bash, zsh, fish, and PowerShell, leveraging Cobra's built-in functionality for this purpose. The acceptance criteria include verifying the output for each shell type and documenting the new feature in the project's README.

Complexity: 2/5
enhancement good first issue

Security scanner for AI agent skills & MCP servers. 173 detection rules. 13 categories. 5 registries monitored daily. OpenClaw detection included. No API keys, no cloud, no LLM. One binary. Detection engine behind Oktsec.

Go
#ai-agents#claude#claude-code#golang#mcp#openclaw#openclaw-skills#prompt-injection#security#skills

AI Summary: This issue requests the addition of a test coverage badge to the project's README. It involves integrating a coverage service like Codecov or Coveralls into the existing CI workflow to upload coverage data, then configuring a coverage threshold, and finally displaying the badge in the README.

Complexity: 2/5
documentation good first issue

Security scanner for AI agent skills & MCP servers. 173 detection rules. 13 categories. 5 registries monitored daily. OpenClaw detection included. No API keys, no cloud, no LLM. One binary. Detection engine behind Oktsec.

Go
#ai-agents#claude#claude-code#golang#mcp#openclaw#openclaw-skills#prompt-injection#security#skills