MrBinnacle
UserAgent skills for Claude Code that have to earn their keep: each carries evidence of the real failure behind it, gets re-tested on new model releases, and is publicly retired when models stop needing it.
Categories
Indexed Skills (7)
closure-mode-at-boundaries
Closure-mode discipline at workflow boundaries. Use when a sprint, phase, or vertical slice locks clean, or when a multi-candidate "what next" decision arises. Owns the closure→build transition.
skill-necessity-gate
Decide whether a capability should become a skill. Use when asking "should this be a skill?", pruning a library, choosing skill vs CLAUDE.md vs MCP vs hook, or before building an eval instrument.
git-pull-rebase-trap
Use before `git pull` where `pull.rebase` may be `true`. `--no-ff` does NOT stop a rebase under `pull.rebase=true` (silently ignored); it rewrites every local commit SHA. Check config first.
github-pages-deploy-verification
Verify a GitHub Pages (or any CDN-fronted) deploy actually serves new content. Poll on a marker that did not exist pre-deploy; avoid sleep-and-curl chains the agent harness blocks.
downstream-instruction-framing
Framing discipline for any artifact instructing a downstream reader — handoffs, plans, ADRs, subagent prompts. Bans blanket do-not-re-litigate framing; requires per-decision Revisit-if clauses.
parallel-review-disposition-schema
Use when dispatching parallel agents to verify/adjudicate findings — adversarial verify, security disposition, design triage, council fires. Shared disposition schema so seat outputs join cleanly.
subagent-research-reliability
Reliability checks for web-research subagents. Pre-dispatch, verify the agent's tool grant actually includes web tools (else it no-ops or fabricates). Post-return, verify every citation before acting.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.