A Claude Code plugin that writes an over-limit prompt to a temp file and sends the model the first lines with the file path, saving context.
✳︎ prompt-offload
└─ Make every conversation easier
A Claude Code plugin that writes an over-limit prompt to a temp file and sends the model the first lines with the file path, saving context.
A Claude Code plugin that learns your frequently used short prompts per project and draws them above the prompt box, sending one with a digit key.
For mod development: runs a live check of plugins in a fresh Claude Code session in tmux, types the steps one by one, collects the pane and the transcript, then deletes the temp directory, its transcripts and the inline store files.
A Claude Code plugin that keeps notes you pin with `/pin-note` for the whole session and re-injects them verbatim into the model's context after each compaction and `/clear`, via a SessionStart hook, with per-session persistence in `$.store` and an off-by-default toggle.
A Claude Code plugin that measures how much context each Bash command consumed and, past a size limit, tells the model which narrower command would have answered the same question.
Lists the servers a Bash call of the model started in this repository and left listening on a port, with their age and session, and stops one with SIGTERM, then SIGKILL.
A Claude Code mod that names installed plugins whose marketplace clone already offers a newer version, and prints the claude plugin update command to close the gap.
A memory kept by a Stop hook blocks the turn's end, makes the model Read and Edit MEMORY.md in front of you, and costs an extra turn every time. This mod keeps a per-project MEMORY.md up to date without blocking the stop, and loads it into the session.
Tells you when an MCP server failed to connect or dropped, with the engine's reason and a reconnect button, and when it is back; claude.ai connectors are left out.
Checks after git commit whether manifest dependency changes left the matching lockfile behind, supporting npm, Composer, Cargo, Go, Python, Bundler, Dart and Mix; optionally blocks commit, push and merge while a lockfile is behind.
A Claude Code mod that keeps subscription usage limits on screen with reset countdowns, pace forecasts, a /limit-watch pane, and 80%/95% warnings.
A Claude Code plugin that draws ASCII animations (matrix rain, fire, aquarium, cat, or imported GIFs) above the prompt while the model works; display-only and costless.
After an edit that calls translation keys, names the keys the JSON, Laravel PHP, YAML or gettext locale files lack, and the languages that lack them.
A Claude Code plugin that ships a git commit skill and gates every Bash git command against it — blocking commits made without the skill, blanket `git add`, AI signature trailers, secrets, ignored paths, and unrequested pushes or branch changes.
A Claude Code plugin that hooks the Bash tool to have Gemini review every git commit the model makes, blocking commits with bugs, data loss, security holes or committed secrets, based on the staged diff and the conversation.
Reviews each plan with Gemini before the approval dialog opens and sends a plan with blocking issues back to the model at most twice.
Keeps the Gemini keys and tier of every Gemini mod, and each mod's model and thinking level, in one place: adds $.gemini, which builds their Gemini requests, reads the answers and moves to the next key after a quota or key error, and picks a model from Google's list.
Moves compaction from Claude to Gemini: in summary mode Gemini summarizes the conversation and the newest messages stay verbatim, in prune mode every message stays and Gemini keeps, cuts or drops older tool calls. Off until /gemini-compact on; gemini-core holds the key, tier, model and thinking level.
A Claude Code plugin that gives the model a self-invoked Gemini advisor tool: it sends the full conversation plus the model's message to Gemini and returns a second opinion. Off by default until /gemini-advisor on; gemini-core holds the key, tier, model and thinking level.
A Claude Code mod that fingerprints the working tree and remembers which tests passed and failed on the same code, so the model re-runs a flaky test instead of changing working code.
A Claude Code plugin that watches main-loop turn.complete events and, when a turn ends with reason "error" (API error, retries exhausted, context limit) or a usage limit, submits one continue prompt so the model resumes half-done work, up to a configurable limit (99 by default).
After each commit the model makes, names the env variables its added lines read that .env.example lacks, with the file and line of each.
A Claude Code plugin that rates each prompt's difficulty with a small model and runs that turn and its subagents at the matching effort level, within user-allowed limits, on models whose prompt cache survives effort changes.
A Claude Code plugin that hooks Edit, Write and NotebookEdit tool calls, counting successful edits per file and turn; at the third edit it warns the user and at the fifth it sends the model a note to stop editing, re-read the code path and state the root cause. Counts reset each turn and are tracked separately for the main loop and subagents.
A Claude Code plugin that hooks into the Bash tool and, after each git commit, uses ripwire to detect markdown anchors (file:line references, symbol names) broken by the commit, feeding stale lines back to the model and holding each finding open until the doc holds again.
A Claude Code plugin that measures git-ignored build artifacts (node_modules, target, .venv, etc.), shows the total on the status line from 5 GB, and deletes only the directories you pick twice in the /disk-janitor pane; data directories are never listed.
A Claude Code mod that renders mermaid blocks from the model's replies with an installed mmdc and draws each picture under its reply.
Sends a desktop notification when a question or a plan waits for your answer, and when a turn ends or fails.
Checks each package the model installs against its registry and OSV.dev, and stops a missing, brand-new, look-alike, outdated or vulnerable one, naming the latest version.
Asks a council of models about a hard problem, when the model calls its tool or you run /council <question>: Claude models, and Gemini models when gemini-core has a key, answer in parallel, and the model of the session writes one verdict from their answers.
After the model changes a function signature with Edit, adds the callers ripwire finds to the Edit's result, so the model fixes them before the build does.
Claude Code mod that re-sends the current text of a skill, command, or rules file to the model when the file changes on disk after the session loaded it.
Parses each JSON, YAML, TOML or .env file an Edit or Write touches, notes a parse error at once instead of at the next build, and says so again when a later edit fixes it.
A Claude Code plugin that names uncommitted files at the end of each turn and reminds the model to commit finished work.
Keeps the 1-hour prompt cache warm for a window you set, or with no end in every session under always, with one cache-shared fork per idle stretch and one keep-warm message after a resume, arms a window for you after a paid cold write, and shows the cache state, the cold price and this session's cold writes.
Runs proof-driven bug hunt rounds with /bughunt: each round proves one bug with a failing command the mod runs itself, fixes it, and proves the fix, and the loop stops on a blocked or unverified round. /bughunt collab runs a read-only scanner, planner and critic in sequence.
A Claude Code plugin that tracks every background shell task in a session, shows the oldest task's age in the status line or sidebar, and lets you stop one from the /bg-tasks pane.
Shrinks each Bash result before the model reads it: per-command filters for git, test runners, linters, package managers, containers and file tools, the full output kept in a file.
Picks the recommended option of a question that waited unanswered for a set time, so a session you left does not stop. Off until you turn it on.
A Claude Code mod that watches Edit/Write edits to GitHub Actions workflows and, when a step is pinned to a moving ref (tag or branch), tells the model the commit SHA to write instead; an optional deny mode blocks git commit, push and merge until refs are pinned.