output-cap 0.1.0
Bash results over 150 lines reach Claude as their first and last lines with a note of how many lines were cut, so a long log or test run does not flood the context.
Install
/plugin marketplace add shonzinger-spec/claude-code-mods /plugin install output-cap@shon-mods
Run both lines inside Claude Code.
Hooks into
- Tool calls
Also in shonzinger-spec/claude-code-mods
- effort-router 0.2.0 Picks the effort level and whether to use ultracode for each prompt you type (Haiku + calibration rules, keyword fallback), with !low..!max, !ultra and !solo overrides. 0
- paste-dedupe 0.1.0 When a prompt repeats a paste from an earlier message this session (8+ identical lines), the repeat is replaced by "[same as message N: K lines …]" so the same log or file is not sent to Claude twice. 0
- cost-band 0.1.0 Status line with session $ and cache ratio ("$12.34 · cache 87%"), always on; above the prompt an amber band from the nudge cost (default $15) saying to start a new session, red at the limit (default $20), and one toast at each. Refreshes after each turn and every 60 s. 0
- session-health 0.1.0 Warns when a session is too worn to trust (context fill, compactions, tool errors) and saves a verified handoff + checks CLAUDE.md before you switch. 0
- profile 0.1.0 /profile: switch which plugin set loads next session (trading, business, coding, all), while preserving a restorable full-plugin snapshot. 0
- sharpen 0.1.0 Turns a rough idea into a clear, complete brief before Claude starts work (auto, ++ prefix, or /sharpen for other AIs). 0
- folder-guard 0.1.0 Refuses a NEW file or folder loose in ~, ~/Desktop or ~/Documents (Write, NotebookEdit, and Bash redirects/touch/mkdir/cp/mv destinations) and suggests ~/Trading, ~/YouTube, ~/AutoRank-Business or ~/Career by keyword. Edits to existing files, existing project folders and dot-folders pass. Quote- ... 0
- long-run 0.1.0 Backtest-like python/uv/poetry runs: remembers how long each command signature took, refuses a foreground re-run of one that took 2+ minutes (asks for run_in_background), and sends a desktop notification when a long run ends, in the foreground or the background. 0
- prereg 0.1.0 /prereg <slug> <plan> has Claude write prereg/<slug>.json (hypothesis, gates, thresholds, SHA256) before any result is seen. A guard then refuses edits to a pre-registration once results exist: prereg/<slug>.result.json or results/<slug>*, and the legacy PREREGISTRATION_<tag>.md once PREREG_<tag>... 0
- claim-check 0.1.0 Flags PF / profit factor / Sharpe / win rate / WR / t-stat numbers in Claude's final reply that no tool result or user prompt this session backs at the stated precision (corpus 1.3127 backs "1.31"). One toast per turn; /claims lists the last reply's metrics, backed or not, with the nearest corpus... 0
More guardrails mods
- blast-radius 0.1.0 See what a risky command would change before it runs. 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... 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 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. 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. 33.1k