Individual skill

self-evolve

Improve an existing skill or repository through isolated proposals, reproducible evaluation and evidence-based acceptance. Use for self-evolve or skill improvement.

SKILL.md

Read pinned source

Source commit 176a9e8aea3a0795eeee51760a4132e52de12be3

What was checked

Matches the observed repository HEAD: Yes
Scanner rules are current: Yes
Repository scan complete: No
Skill scan complete: No
Observed behavior: Network or broad shell
Last checked: Nov 4, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings): 26
Repository last updated: Oct 16, 2026
License: MIT

Observed agent integration: claude-code

agent Repository topic
agent-evolution Repository topic
ai Repository topic
ai-agent Repository topic
ai-safety Repository topic
autonomous-agents Repository topic
claude Repository topic
claude-code Repository topic
claude-plugin Repository topic
claude-skill Repository topic
engineering Derived category

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files

.claude-plugin/plugin.jsonasset · 837 bytes
.dataclass.jsonasset · 1,747 bytes
.githooks/pre-commitruns · 628 bytes
.githooks/pre-pushruns · 337 bytes
.gitignoreasset · 441 bytes
.pii-allowasset · 599 bytes
CHANGELOG.mddoc · 7,108 bytes
DATA.mddoc · 11,866 bytes
LICENSEasset · 1,066 bytes
PHILOSOPHY.mddoc · 6,947 bytes
README.mddoc · 7,336 bytes
README_CN.mddoc · 6,758 bytes
ROADMAP.mddoc · 9,584 bytes
SKILL.mdcore · 3,377 bytes
SKILL.mdcore · 3,377 bytes
commands/self-evolve-resume.mddoc · 862 bytes
commands/self-evolve-status.mddoc · 750 bytes
commands/self-evolve.mddoc · 897 bytes
config.contract.jsonasset · 645 bytes
docs/diagrams/render.pyruns · 2,424 bytes
docs/diagrams/workflow-cn.dotasset · 1,859 bytes
docs/diagrams/workflow-cn.pngasset · 113,954 bytes
docs/diagrams/workflow-en.dotasset · 1,772 bytes
docs/diagrams/workflow-en.pngasset · 96,884 bytes
examples/btarget/report.jsonasset · 15,103 bytes
reference/acceptor_math.mddoc · 8,841 bytes
reference/evaluation.mddoc · 7,481 bytes
reference/maintenance.mddoc · 3,914 bytes
reference/runtime.mddoc · 10,042 bytes
requirements.txtdoc · 277 bytes
retention.json.exampleasset · 135 bytes
schemas/history-catalog.schema.jsonasset · 3,600 bytes
schemas/storage-manifest.schema.jsonasset · 2,599 bytes
schemas/storage-retention.schema.jsonasset · 1,210 bytes
scripts/setup_btarget_repo.pyruns · 6,013 bytes
scripts/validate_btarget.pyruns · 5,232 bytes
spikes/__init__.pyruns · 0 bytes
spikes/confseq_spike.pyruns · 2,617 bytes
storage.contract.jsonasset · 45,593 bytes
tools/__init__.pyruns · 0 bytes
tools/llm_fixtures.pyruns · 5,983 bytes
tools/make_fixtures.pyruns · 70,826 bytes
tools/sie/__init__.pyruns · 38 bytes
tools/sie/acceptor.pyruns · 16,491 bytes
tools/sie/agents.pyruns · 3,540 bytes
tools/sie/anchors.pyruns · 9,886 bytes
tools/sie/archive.pyruns · 12,049 bytes
tools/sie/backends/__init__.pyruns · 0 bytes
tools/sie/backends/builtin.pyruns · 570 bytes
tools/sie/backends/llm.pyruns · 17,795 bytes
tools/sie/business_tree.pyruns · 10,051 bytes
tools/sie/calibrate.pyruns · 39,784 bytes
tools/sie/calibration_defects/cl-oss-lane-key-mismatch.jsonasset · 2,182 bytes
tools/sie/calibration_defects/cl-reward-floor-flattens-low-end.jsonasset · 2,873 bytes
tools/sie/calibration_defects/cl-track-weight-double-charged.jsonasset · 4,117 bytes
tools/sie/calibration_defects/cl-unseen-arm-inherits-best.jsonasset · 2,814 bytes
tools/sie/calibration_defects/cp-report-failure-collapses-into-cannot-check.jsonasset · 3,528 bytes
tools/sie/calibration_defects/cp-unparseable-digest-name-read-as-a-clean-day.jsonasset · 3,192 bytes
tools/sie/calibration_defects/cp-unscanned-archive-publishes-zero-missing-days.jsonasset · 6,887 bytes
tools/sie/calibration_defects/lb-demand-bar-measured-against-the-shipped-floor.jsonasset · 2,997 bytes
tools/sie/calibration_defects/lb-origin-labels-lose-case-folding.jsonasset · 2,526 bytes
tools/sie/calibration_defects/lb-quote-capped-at-the-push-limit.jsonasset · 2,480 bytes
tools/sie/calibration_defects/lb-undated-age-scores-as-brand-new.jsonasset · 2,097 bytes
tools/sie/calibration_defects/sc-age-floored-at-collection-cadence.jsonasset · 3,132 bytes
tools/sie/calibration_defects/sc-crowdedness-damper-outside-the-weights.jsonasset · 3,855 bytes
tools/sie/calibration_defects/sc-final-map-drops-lifecycle-stage.jsonasset · 3,334 bytes
tools/sie/calibration_defects/sc-undated-evidence-counts-toward-the-red-line.jsonasset · 3,185 bytes
tools/sie/calibration_defects/yl-lane-alias-not-folded.jsonasset · 3,231 bytes
tools/sie/calibration_defects/yl-previral-key-list-drift.jsonasset · 4,050 bytes
tools/sie/calibration_defects/yl-propose-add-silent-cap.jsonasset · 3,312 bytes
tools/sie/calibration_defects/yl-unparsed-ts-reads-as-in-window.jsonasset · 3,843 bytes
tools/sie/check_reflection.pyruns · 3,063 bytes
tools/sie/cli.pyruns · 10,766 bytes
tools/sie/edgar_cache.pyruns · 1,079 bytes
tools/sie/evaluate.pyruns · 18,877 bytes
tools/sie/events.pyruns · 3,154 bytes
tools/sie/gate_human.pyruns · 2,931 bytes
tools/sie/immutable.pyruns · 7,037 bytes
tools/sie/judge_claude.pyruns · 392 bytes
tools/sie/judge_codex.pyruns · 389 bytes
tools/sie/judges.pyruns · 10,126 bytes
tools/sie/llm_adapter.pyruns · 6,603 bytes
tools/sie/llm_agent_child.pyruns · 962 bytes
tools/sie/patch.pyruns · 44,812 bytes
tools/sie/probes/__init__.pyruns · 0 bytes
tools/sie/probes/exec_probe.pyruns · 6,670 bytes
tools/sie/probes/fact_probe.pyruns · 3,770 bytes
tools/sie/profile.pyruns · 10,224 bytes
tools/sie/propose.pyruns · 1,673 bytes
tools/sie/proxy.pyruns · 11,123 bytes
tools/sie/reflect.pyruns · 5,135 bytes
tools/sie/runtime_data.pyruns · 15,036 bytes
tools/sie/sandbox.pyruns · 8,632 bytes
tools/sie/selfboot.pyruns · 4,202 bytes
tools/sie/selfdeception.pyruns · 7,796 bytes
tools/sie/stage_records.pyruns · 3,028 bytes
tools/sie/state.pyruns · 840 bytes
tools/sie/statemachine.pyruns · 74,676 bytes
tools/sie/storage.pyruns · 5,847 bytes
tools/sie/supervisor.pyruns · 8,222 bytes
tools/sie/verifiable.pyruns · 18,786 bytes
tools/sie_cli.pyruns · 265 bytes
tools/storage_retention.pyruns · 8,613 bytes
workflows/_agent_launch.jsruns · 434 bytes
workflows/_claude_launch.jsruns · 434 bytes
workflows/agent.jsruns · 434 bytes
workflows/claude-judge.jsruns · 434 bytes
workflows/claude-propose-artifact.jsruns · 434 bytes
workflows/claude-propose.jsruns · 434 bytes
workflows/codex-judge.jsruns · 434 bytes
workflows/reflect-fanout.jsruns · 434 bytes
workflows/review-fanout.jsruns · 434 bytes

Read excerpt


# self-evolve

在 Git worktree 中提出修改、运行评测，再按证据决定是否采纳。
先明确目标和可执行判据；通过检查只证明所测范围。

## 开始前

1. 确认目标仓库、基线 revision、允许修改的范围和期望改善的行为。
   记录文档影响；无需修改的文档也说明原因。设计理由见 [PHILOSOPHY.md](PHILOSOPHY.md)。
2. 确定评测信号：A 为测试结果，B 为独立事实核验，C 为已提供的主观证据。
   主循环尚无 C 回归与一致性测量生产者，也不执行 A+B 组合。
3. 按 [DATA.md](DATA.md) 配置 PRIVATE 伴生仓，保留真实运行证据。
4. 运行 `python <skill-path>/tools/sie_cli.py doctor --target <target>`，
   阅读配置和能力缺口。doctor 不调用模型，退出零不能单独证明就绪。