ouroboros-loops

Plugin ouroboros-loops from marketplace jylee-claude-marketplace in JuyeongYi/jylee_claude_marketplace. Commands read from the repo's manifest — 1 month ago.

Skills (12)

• ouroboros-loops:analyze

  Analysis stage of the Ouroboros loop. Reads raw observations from the research step (.ouroboros/research.md) and classifies them into bugs, quality_issues, and enhancement_opportunities. Saves results to .ouroboros/state.json.

• ouroboros-loops:branch

  Large-scope task isolation. Detects oversized tasks from analyze/fix, creates a git branch, optionally launches a background agent, and records the branch plan for the main loop to track.

• ouroboros-loops:cancel

  This skill should be used when the user asks to 'cancel ouroboros', 'stop ouroboros loop', 'stop the loop', 'ouroboros cancel', or 'halt autonomous improvement'. Gracefully cancels an active Ouroboros loop.

• ouroboros-loops:context-handoff

  Saves a handoff summary to .ouroboros/handoff.md after each Ouroboros step completes. Called by the ouroboros-agent after invoking a step skill, so the next step agent can start with compressed context instead of a full history.

• ouroboros-loops:determine

  Post-evaluate autonomous judgment stage for the Ouroboros loop. Performs a comprehensive assessment of codebase health, stagnation detection, and natural-language evaluation of the completionPrompt condition. Returns OUROBOROS_RESULT: done or continue. Runs after evaluate in the step sequence.

• ouroboros-loops:enhance

  Third stage of the Ouroboros loop. Select realistic enhancement opportunities from the analyze results and implement them in this cycle.

• ouroboros-loops:evaluate

  Final stage of the Ouroboros loop. Check completion conditions (iteration count, prompt match, or no remaining issues) and return 'continue' or 'done' to the agent. On continue, also increments the iteration counter in state.json.

• ouroboros-loops:fix

  Second stage of the Ouroboros loop. Select the highest-impact issues from the analyze results and apply targeted fixes in this cycle.

• ouroboros-loops:report

  Fourth stage of the Ouroboros loop. Save a markdown report of this iteration's work and create a git commit.

• ouroboros-loops:research

  Pre-analysis stage of the Ouroboros loop. Scans the codebase structure, reads key files, and optionally performs web searches for external context (known issues, CVEs, best practices). Saves raw observations to .ouroboros/research.md for the analyze step.

• ouroboros-loops:resume

  This skill should be used when the user asks to 'resume ouroboros', 'continue ouroboros loop', 'ouroboros resume', 'continue the loop', 'loop stopped', or 'restart ouroboros'. Resumes an interrupted Ouroboros loop from where it left off without losing progress.

• ouroboros-loops:start

  This skill should be used when the user asks to 'start ouroboros', 'run ouroboros loop', 'start autonomous improvement', 'begin ouroboros', or 'ouroboros start'. Starts an autonomous codebase improvement loop that iterates through research → analyze → fix → enhance → report → evaluate cycles.

Subagents (1)

• ouroboros-agent

  Ouroboros single-step executor. Invoked once per step by the ouroboros-loops:start or ouroboros-loops:resume skill. Reads handoff context from the previous step, executes the specified step skill, then saves a handoff for the next step. Model is specified by the caller (Opus for analyze/evaluate, So

Hooks (5)

• Stop

  ${CLAUDE_PLUGIN_ROOT}/hooks/ouroboros-watchdog.sh

• TeammateIdle

  ${CLAUDE_PLUGIN_ROOT}/hooks/teammate-idle-quality-gate.sh

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/hooks/mcp-pre-tool-log.sh

• PostToolUse

  ${CLAUDE_PLUGIN_ROOT}/hooks/mcp-post-tool-log.sh

• PostCompact

  ${CLAUDE_PLUGIN_ROOT}/hooks/post-compact-reinject.sh

Similar plugins