Plugin Claude Code
Plugin marketplace listing 1 plugin: loopy.
1 1mo ago A
tokens not measured
original MIT
loopy
02Plugin Claude Code
Universal loop-engineering harness: machine-verifiable stop conditions, cross-model maker/checker (Codex maker, Claude checker), disk-based loop memory, reversibility×impact decision gates that keep human approval to irreversible actions while auto-pushing reversible work-branch commits, autonomous replan on stalls, a.
1 1mo ago A
tokens not measured
original MIT
Instructions file
Claude Code instructions for hhamja/claude-code-harness, a project described as: Claude Code plugin for loop engineering: implement-verify cycles with machine-verifiable stop conditions, an independent read-only verifier agent, and disk-based state that survives session death.
1 1mo ago A 289 tokens
original MIT
auditor
04Agent
Read-only auditor of loop process and principles. Grades maker/checker separation, machine-verifiable stops, disk memory, and gate adherence; returns a scored report. Never modifies files.
1 1mo ago A 39 tokens
original MIT
Agent
Adversarial red-team reviewer of a loop/harness — tries to REFUTE compliance claims (gate bypasses, forgeable approvals, rubber-stamps, script bugs) and reports reproduced holes. Never modifies files.
1 1mo ago A 47 tokens
original MIT
explorer
06Agent
Low-cost read-only codebase scout for loop tasks. Maps relevant files and returns a compact findings report.
1 1mo ago A 24 tokens
original MIT
Agent
Read-only loop/harness architect grounded in the control plane. Diagnoses a target project's ETCLOVG coverage and returns a scored report with a maturity level. Never modifies files.
1 1mo ago A 39 tokens
original MIT
verifier
08Agent
Read-only grader for loop cycles. Checks work against .claude/loop/rubric.md only and returns a per-criterion pass/fail report with evidence. Never modifies files.
1 1mo ago A 40 tokens
original MIT
SessionStart
09Hook
Runs when a session starts on startup, resume, clear and compact, executing session_state_digest.sh and loop_reminder.sh (3 commands). From hhamja/claude-code-harness.
1 1mo ago A
tokens not measured
copy · 64% MIT
PreToolUse
10Hook
Runs before the agent uses a tool for Bash tool calls, executing verifier_guard.sh and decision_gate.sh (2 commands). From hhamja/claude-code-harness.
1 1mo ago A
tokens not measured
copy · 66% MIT
PostToolUse
11Hook
Runs after a tool call finishes for Bash, Edit, Write and NotebookEdit tool calls, executing touch_track.sh. From hhamja/claude-code-harness.
1 1mo ago A
tokens not measured
original MIT
Stop
12Hook
Runs when the agent finishes a response, executing stop_gate.sh, check_memory.sh, auto_commit.sh, auto_push.sh and auto_pr.sh (5 commands). From hhamja/claude-code-harness.
1 1mo ago A
tokens not measured
copy · 59% MIT
loop-audit
13Skill Claude CodeCodex
Audit the loop against loop-engineering principles and decision-gate adherence; write a scored report to audit.md.
1 1mo ago A 25 tokens
original MIT
loop-ci
14Skill Claude CodeCodex
Scaffold a GitHub Actions CI workflow from the loop's detected test/lint/build commands.
1 1mo ago A 21 tokens
original MIT
Skill Claude CodeCodex
Loop engineering method — machine-verifiable stop conditions, maker/checker separation (Codex CLI or Claude as maker, a verifier subagent as checker), disk-based memory in .claude/loop/. Consult when designing or running agent loops.
1 1mo ago A 52 tokens
original MIT
loop-init
16Skill Claude CodeCodex
Scaffold .claude/loop/ state files and detect this project's test/lint/build commands.
1 1mo ago A 22 tokens
original MIT
loop-review
17Skill Claude CodeCodex
Comprehensively review a project's loop/harness engineering against the control plane — ETCLOVG coverage, maturity, adversarial red-team — writing a scored harness-review.md; with --fix, remediate the high-severity findings on a branch via loop-run and stop at a PR.
1 1mo ago A 59 tokens
original MIT
loop-run
18Skill Claude CodeCodex
Run maker/checker cycles — implement via Codex CLI or Claude per loop.config.md, verifier grades — until the rubric passes; supports --once and --verify-only.
1 1mo ago A 36 tokens
original MIT
loop-status
19Skill Claude CodeCodex
Summarize loop state, rubric progress, and the last run token estimate from .claude/loop/.
1 1mo ago A 25 tokens
original MIT
Skill Claude CodeCodex
Spawn an isolated git-worktree worker for a rubric subset; integrate finished workers back.
1 1mo ago A 20 tokens
original MIT