메뉴
Inspect and operate OV5 (Ouroboros V5) through the live Emacs daemon. Use when checking auto-workflow status, starting guarded runs, reviewing experiment results, or querying researcher and evolution state.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Transforms research paper analysis from extraction to narrative storytelling. Uses 7-beat narrative spine (protagonist/dilemma/old-path/turning-point/solution/ending/core) to make papers understandable to non-experts. Includes speed-read card, PhD advisor review, and real-world testing. Use when researcher encounters a research paper and needs to extract deep understanding, not just surface facts. Triggers on "paper", "research paper", "analyze paper", "tell me about this paper", "讲论文", "读论文".
Workflow chain: researcher → paper-storytelling. When researcher encounters a research paper, automatically invoke paper-storytelling to transform extraction into narrative understanding. Use when user says "research paper", "analyze this paper", "tell me about this paper", or when researcher detects arxiv/PDF links.
Prompt template for external research specialist subagent. Auto-evolves based on experiment outcomes.
Clojure REPL client (nREPL-based, Babashka). Use for evaluating Clojure code, loading Clojure files, fixing unbalanced brackets, and interactive nREPL work. Not the Elisp daemon-repl.
Daemon REPL for Elisp — evaluate Elisp code in a running Emacs daemon via emacsclient, validate brackets before save, auto-evaluate .el files on change. Use when you need to run Elisp from outside Emacs, check daemon status, or validate Elisp syntax.
Orchestrates automated code improvement through hypothesis-driven experimentation and self-evolution
| name | ov5 |
| description | Inspect and operate OV5 (Ouroboros V5) through the live Emacs daemon. Use when checking auto-workflow status, starting guarded runs, reviewing experiment results, or querying researcher and evolution state. |
OV5 is the live Ouroboros V5 self-improving system behind the pmf-value-stream Emacs daemon.
Subsystems:
auto-workflow): select -> categorize -> route -> hypothesis -> tests -> grade -> review -> merge/learngh API and refreshes research cacheself-evolve): generates new strategies from failure patternsCurrent architecture facts from OUROBOROS-V5.md: 120 modules (76 byte-compiled, 44 no-byte-compile), 3,485 ERT tests, 7 gates, 8 backend definitions (4-5 actively routed), ~20% keep-rate, ~59% prompt compression, and ~$0.50-2.00/run.
Use the main OV5 daemon: pmf-value-stream.
Preferred socket path:
/tmp/emacs$(id -u)/pmf-value-stream
Resolution order:
$XDG_RUNTIME_DIR/emacs/pmf-value-stream${TMPDIR}/emacs$(id -u)/pmf-value-stream/tmp/emacs$(id -u)/pmf-value-streamQuick check:
ls "/tmp/emacs$(id -u)/"
emacsclient patternAlways use heredoc plus -a false for OV5 calls. See also the daemon-repl skill for shared emacsclient guidance.
emacsclient -s /tmp/emacs$(id -u)/pmf-value-stream -a false --eval "$(cat <<'EOF'
(gptel-auto-workflow-status)
EOF
)"
Rules:
-a false so emacsclient never auto-spawns a daemon.progn.| Form | Returns / use | Notes |
|---|---|---|
(gptel-auto-workflow-status) | (:running :kept :total :phase :run-id :results) | Main status entrypoint. |
(gptel-auto-workflow-run-async) | Starts a run | Raw async start; returns 'started. |
(gptel-auto-workflow-run-async--guarded) | Starts a guarded run | Preferred for automation. |
(gptel-auto-workflow-log) | Last 20 sanitized log lines | Filters [auto-] and [nucleus] messages. |
(gptel-auto-workflow-read-persisted-status) | Status from disk | Survives daemon restarts. |
(gptel-auto-workflow-research-status) | Researcher status plist | Cache freshness and timer state. |
(gptel-auto-workflow-run-research) | Refreshes research cache | Public researcher entrypoint. |
(gptel-auto-workflow-evolution-status) | Evolution subsystem status | Buffer-oriented / interactive. |
Use these with bound-and-true-p:
(bound-and-true-p gptel-auto-workflow--running)(bound-and-true-p gptel-auto-workflow--current-target)emacsclient -s /tmp/emacs$(id -u)/pmf-value-stream -a false --eval "$(cat <<'EOF'
(gptel-auto-workflow-status)
EOF
)"
emacsclient -s /tmp/emacs$(id -u)/pmf-value-stream -a false --eval "$(cat <<'EOF'
(gptel-auto-workflow-run-async--guarded)
EOF
)"
If it returns nil, check for pending decisions in mementum/decisions/ or an already-running workflow.
emacsclient -s /tmp/emacs$(id -u)/pmf-value-stream -a false --eval "$(cat <<'EOF'
(gptel-auto-workflow-log)
EOF
)"
Then inspect:
var/tmp/cron/auto-workflow-status.sexpvar/tmp/experiments/*/results.tsvgit log --oneline -10Auto-discovered from lisp/modules/ (skips -test.el, -disabled.el, /test/). Override in .dir-locals.el:
((emacs-lisp-mode . ((gptel-auto-workflow-targets . ("lisp/modules/foo.el")))))
emacsclient -s /tmp/emacs$(id -u)/pmf-value-stream -a false --eval "$(cat <<'EOF'
(gptel-auto-workflow-run-research)
EOF
)"
gptel-auto-workflow--rate-limited-backends function exists.gptel-auto-workflow--read-analysis function exists.gptel-auto-workflow-strategic-research function exists.| Path | Purpose |
|---|---|
var/tmp/cron/auto-workflow-status.sexp | Persisted workflow status |
var/tmp/cron/auto-workflow-messages-tail.txt | Persisted message tail |
var/tmp/experiments/*/results.tsv | Per-run experiment results |
var/tmp/experiments/*/optimize/ | Per-experiment worktree output |
var/tmp/research-findings.edn | Researcher cache |
var/tmp/cross-subsystem-state.json | Cross-subsystem persisted state |
var/approval-queue/pending/ | Pending high-risk proposals |
var/approval-queue/decisions/ | Approved/rejected/expired proposals |
mementum/decisions/ | Human decision gate |
0 10,14,18 * * * pipeline (3x daily)
0 2 * * * mementum
0 3 * * * instincts
-a false.gptel-auto-workflow-run-async--guarded for agent-driven starts.main directly — isolated git worktrees only.daemon-repl — shared emacsclient/heredoc patterns, Elisp evaluationbrepl — Clojure nREPL client (separate tool)ov5-status — focused system health checkov5-handover — session handover for OV5