marker-guard 0.1.0
Enforces AI-01 for the user's benefit. The [CX] marker exists so the user can see that the AI has not forgotten the project context. When a main-loop answer does not start with the marker the project's CLAUDE.md or AGENTS.md defines (with its Context code, if one is set), this mod shows one warni...
Aftab Ibrahim KaziInstall
git clone https://github.com/AftabIbrahimKazi/ai-dev-kit.git claude --plugin-dir ai-dev-kit/mods/marker-guard
Run in a terminal. Claude Code loads the mod from that folder for the session.
Hooks into
Nothing we recognise automatically. See the source on GitHub.
Also in AftabIbrahimKazi/ai-dev-kit
-
aiignore-guard
0.5.0
Keeps the AI out of files it should not read, in four layers: the project's .aiignore (always), paths that belong to another role (the active role's "out" list, or any path another role lists under "in" that the active role does not; .claude/conflict-roles.json; the role comes from AI_ROLE, a pro...
Aftab Ibrahim Kazi in Guardrails, updated
9
-
guard
0.1.0
Denies edits to secret files, lockfiles, .git and node_modules, and destructive Bash or PowerShell commands (recursive deletes of root-like paths, force push, deleting remote main/master, git reset --hard, git clean -f, publish, a download piped into a shell, SQL drops). Per-project config in .cl...
Aftab Ibrahim Kazi in Guardrails, updated
8
-
prefilter-rank
0.1.0
After a Grep/Glob with 5+ results, ranks them by relevance to the task using the local Laya prefilter (system1-prefilter addon) and adds a read-order reminder. Calls Laya locally at 127.0.0.1 only (the host is checked in code, never trusted from the env var alone) -- no other network access, no k...
Aftab Ibrahim Kazi in Guardrails, updated
8
More guardrails mods
-
blast-radius
0.1.0
See what a risky command would change before it runs.
deepseek-ai in Guardrails, updated
269.9k
-
agents-md
0.1.0
AGENTS.md as project instructions, by one option. Under claude-md-or-agents-md a project with no instruction files of its own gets its AGENTS.md files instead: a prompt.context hook asks $.fs.ancestors for every AGENTS.md and .claude/AGENTS.md from the filesystem root down to the working director...
Anthropic in Guardrails, updated
154.3k
-
ruflo-rvf
0.3.7
mod + skill
RVF format for portable agent memory, session persistence, and cross-platform transfer. As a mod (ADR-445): a write guard that keeps secrets out of memory, sessions and transfers, /rvf-mod, and a status file the console shows
ruvnet in Guardrails, updated
80.7k
-
openviking-memory
0.7.21
mod + skill
Long-term semantic memory for Claude Code, powered by OpenViking. Auto-recall relevant memories at session start and capture important information during conversations.
OpenViking in Guardrails, updated
41.4k
-
tool-timing-badge
0.2.0
Times every tool call and wraps the engine's ToolUse rendering with a colored duration badge from the surface's own element table, on the terminal, Desktop and mobile. Hooks tool.call and ui.render using JSX.
claude-code-templates in Guardrails, updated
33.1k