← all publishers

dhruba-datta

@dhruba-datta source repo

Report this profile Claim this profile

• 10 skills
• 0 followers
• 14 hours ago last updated

1. ▌
   Onboard · dhruba-datta bundle

   First-run walkthrough of the claude-papercuts plugin. Use this skill when the user runs /claude-papercuts:onboard, asks what claude-papercuts does, says they're new to the plugin, asks which papercut skill they should enable first, or wants the curated install order. Runs recommend.py to detect which of the nine papercut skills are already installed, picks the next one to enable, and explains why — opinionated ordering, not exhaustive list.
   0
   installs
2. ▌
   Unclear · dhruba-datta bundle

   Restore Claude Code conversation context after an accidental /clear. Use this skill when the user types /claude-papercuts:unclear, mentions accidentally clearing the conversation, asks to undo /clear, or wants to recover lost context from a recent session. Reads the most recent automatic snapshot from .papercuts/snapshots/ and reconstructs what was being discussed.
   0
   installs
3. ▌
   Safe Shell · dhruba-datta bundle

   Block destructive Bash commands before Claude executes them — even in --dangerously-skip-permissions (YOLO) mode where Anthropic's own permission prompts are bypassed. Use this skill when the user runs /claude-papercuts:safe-shell, asks what safe-shell blocks, or wants to know why a command they expected was refused. A PreToolUse hook scans every Bash tool call against a list of irreversible patterns (rm -rf against / or ~, git push --force, git reset --hard HEAD~, mkfs, dd to /dev/sda, curl-pipe-bash, fork bombs) and refuses them with a structured explanation. Block decisions are visible to Claude so it can re-plan.
   0
   installs
4. ▌
   Amnesia Fix · dhruba-datta bundle

   Persistent cross-session memory for Claude Code. Use this skill when the user wants to see their session journal, asks what was worked on in prior sessions, mentions Claude forgetting context, or runs /claude-papercuts:amnesia-fix. A Stop hook silently appends a compact entry (files touched, decisions, next steps, blockers) to .papercuts/journal.md after every turn; a SessionStart hook injects the last 3 entries as context at the top of every session — survives /clear and /compact.
   0
   installs
5. ▌
   Done Prover · dhruba-datta bundle

   Verify Claude's claim that tests are passing. Use this skill when the user wants to know whether a recent "all tests pass" or "feature complete" statement was honest, or when the user mentions Claude lied about completion. Reads the most recent test output from the conversation transcript and surfaces any discrepancy between what Claude claimed and what the test runner actually reported. Auto-triggered by a Stop hook whenever Claude makes a test-pass claim in its final message.
   0
   installs
6. ▌
   Token X Ray · dhruba-datta bundle

   X-ray every source of auto-injected context Claude Code loads at session start — MCP servers, CLAUDE.md files, skills, subagents, and slash commands — and report estimated tokens per source. Use this skill when the user runs /claude-papercuts:token-x-ray, asks where their context tokens went, says Claude feels slow or expensive, mentions /context showing surprising numbers, or wants to know what to disable. Runs the audit.py script in this skill's directory and presents the result verbatim.
   0
   installs
7. ▌
   Skill Budget · dhruba-datta bundle

   Audit installed Claude Code skills against the system-prompt char budget and show which ones may have silently fallen out of the model's visible list. Use this skill when the user runs /claude-papercuts:skill-budget, asks why their skills stopped triggering, mentions installed-but-invisible skills, or wants a Monday-morning health check of their Claude setup. Runs the audit.py script in this skill's directory and presents the result.
   0
   installs
8. ▌
   Skill Doctor · dhruba-datta bundle

   Diagnose why a SKILL.md isn't auto-invoking. Use this skill when the user runs /claude-papercuts:skill-doctor, asks why their skill won't trigger, mentions a skill that Claude keeps ignoring, or wants a lint pass on a SKILL.md they're writing. Runs lint.py to flag missing trigger phrases, vague descriptions, descriptions that overlap with built-in tool training (the root cause of issue
   0
   installs
9. ▌
   Compact Guard · dhruba-datta bundle

   Survive /compact and auto-compact without losing the plan. Use this skill when the user runs /claude-papercuts:compact-guard, asks why Claude lost their plan after a compact, mentions plan-mode state vanishing mid-session, or wants to see the latest pre-compact snapshot. A PreCompact hook silently captures the current task, active todos, files edited, and the last assistant plan to .papercuts/compact-snapshots/ right before compaction runs; a SessionStart hook (only when source=compact) injects that snapshot back as context so the post-compact session knows what it was doing.
   0
   installs
10. ▌
    Subagent Broker · dhruba-datta bundle

    Battle-tested Task-tool delegation templates that survive the subagent-delegation pitfalls reported in issues
    0
    installs