← 返回 Skills 市场
229
总下载
0
收藏
0
当前安装
24
版本数
在 OpenClaw 中安装
/install skillminer
功能描述
Suggest reusable skills from recurring patterns in local memory files. Human review gate, drafts only to skills/_pending/, local-first runner with optional e...
安全使用建议
This skill appears to do what it says: scan your local OpenClaw memory files, surface recurring patterns, and draft candidate SKILL.md files into skills/_pending/ for your review. Before installing or enabling scheduled runs: 1) run a manual scan (scripts/run-nightly-scan.sh) and inspect state/review/ and state/logs/ to confirm behavior; 2) review scripts/lib/secret-patterns.tsv and scripts/lib/secret-scrub.sh so you understand what is redacted — do not rely solely on scrubbing for sensitive data protection; 3) keep FORGE_RUNNER unset unless you explicitly accept that prompt data will be sent to Anthropic (the default is local-only); and 4) check that the required binaries (openclaw, jq, git, flock, bash) are present and that you are comfortable allowing the skill to read your memory/ directory. The only minor inconsistency is the metadata's inclusion of 'claude' as an allowed subprocess while capabilities.network is false — this is explainable by the documented opt-in fallback but worth noting.
功能分析
Type: OpenClaw Skill
Name: skillminer
Version: 0.5.3
The skillminer bundle is a well-architected tool designed to analyze local memory files and suggest reusable skills. It demonstrates significant security awareness, incorporating features such as atomic file writes with validation, flock-based locking to prevent race conditions, and strict regex-based slug validation to prevent path traversal (scripts/lib/slug-validate.sh). Notably, it includes a secret scrubbing utility (scripts/lib/secret-scrub.sh) that redacts sensitive patterns like API keys and PEM blocks before data is persisted. The core logic prompts (prompts/nightly-scan.md) explicitly instruct the AI agent to treat memory files as untrusted data and ignore any instructions contained within them, providing a robust defense against prompt injection from the analyzed history. No malicious behavior or unauthorized data exfiltration was detected.
能力标签
能力评估
Purpose & Capability
The name/description, scripts, and prompts consistently implement a local memory scanner + draft writer that writes to state/ and skills/_pending/. Required binaries (bash, jq, git, openclaw, flock, date) match the implementation. One small mismatch: metadata lists subprocess 'claude' while capabilities.network is set to false — the package supports an optional FORGE_RUNNER=claude fallback that will send prompt data off-host if explicitly enabled by the user; this is documented, but the capability block could more clearly reflect optional network egress.
Instruction Scope
SKILL.md and the included prompts/scripts limit operations to reading memory/ and writing state/ and skills/_pending/. Prompts explicitly forbid executing memory contents, auto-activation, and sending notifications from inside the prompts. The wrapper-and-prompt model is explicit and scoped; the scripts perform locking, atomic writes, slug validation, and secret-scrubbing before persistence.
Install Mechanism
No install spec is declared (instruction-only in registry) and the repository contains shell scripts and a setup.sh. No remote downloads or obscure third-party install URLs are present in the manifest. The README suggests cloning from GitHub for manual install, which is a normal, low-risk pattern.
Credentials
The skill requires no environment credentials and declares no required env vars. It reads local memory files (by design), which may contain secrets; the project includes conservative redaction via scripts/lib/secret-scrub.sh and externalized patterns. Users should note scrubbing is defense-in-depth, not a guarantee. Optional runtime choice FORGE_RUNNER=claude will send prompt data to Anthropic if enabled; this is opt-in and documented, but it is the only path that would cause networked credential exposure.
Persistence & Privilege
always:false and no system-wide permanent privileges requested. The skill writes to its own workspace paths (state/, skills/_pending/, skills/_rejected/) and uses flock-based locking. It does not attempt to modify other skills or global agent configuration. Nested agent invocation (openclaw by default) is expected for this functionality.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install skillminer - 安装完成后，直接呼叫该 Skill 的名称或使用
/skillminer触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v0.5.3
skillminer 0.5.3
- Updated SKILL.md with clearer metadata, including capabilities, network policy, I/O details, and runner explanation.
- Added optional emoji field to SKILL.md.
- Improved metadata structure for environments, binaries, configs, and subprocesses.
- No functionality changes; documentation and manifest enhancements only.
v0.5.2
skillminer 0.5.2
- Improved shell script reliability and error handling.
- Updated documentation (README.md, SKILL.md, CHANGELOG.md) for clarity and usage details.
- Enhanced state and ledger management in scripts.
- Refined prompt handling in skill-writer workflow.
- Minor code and documentation cleanups.
v0.5.1
0.5.1 hotfix: schema-0.4 template fix (CRIT), rollback symmetry in nightly-scan, _pending staging for atomic writes, migrate-state flock
v0.5.0
skillminer 0.5.0
- Bump version to 0.5.0 in SKILL.md.
- No feature or documentation changes aside from the version update.
v0.4.5
skillminer 0.4.5
- Added scripts for secret-scrubbing (secret-scrub.sh) and ledger state migration (migrate-state.sh)
- Enhanced security by including secret scrubbing and state migration utilities
- Miscellaneous updates to documentation and skill metadata
- Updated shell scripts to support new utilities and maintenance workflows
v0.4.4
skillminer 0.4.4
- Shortened and clarified documentation in SKILL.md for easier onboarding.
- Improved trust and security sections for concise explanation.
- No functional/code changes; documentation and usage clarity updates only.
- User guide and onboarding instructions now reference README and USER_GUIDE for full details.
v0.4.3
skillminer 0.4.3
- Updated documentation in SKILL.md and README.md for clarity and detail.
- Minor housekeeping and metadata maintenance.
- No functional or interface changes; this is a documentation update release.
v0.4.2
skillminer 0.4.2
- Added slug validation logic via a new helper script (`slug-validate.sh`) for more reliable command processing.
- Hardened handling of ledger and promotion scripts to improve robustness and error checking.
- Improved prompt templates for nightly scans and skill writing for better clarity and performance.
- Updated documentation to reflect new commands and enhancements.
- Various bug fixes and internal refactoring for stability.
v0.4.1
skillminer 0.4.1
- Documentation update: SKILL.md revised, no code or behavior changes.
- All usage, environment, and flow instructions remain unchanged—this is a maintenance release updating docs only.
v0.4.0
- Added wrapper-dispatch cron prompt files for scheduled execution of bash wrappers.
- Cron jobs now use hardened bash wrappers for all automated scans and writes.
- Updated documentation (README, USER_GUIDE, SKILL.md) to reflect new scheduling and wrapper usage.
- Improved production hardening notes to highlight safer scheduling and execution.
v0.3.3
skillminer 0.3.3
- Documentation updates in SKILL.md for clarity and completeness.
- No functional code changes; release focuses on improved documentation and guidance.
v0.3.2
Production hardening and reliability improvements.
- Adds atomic write and wrapper validation for state.json to prevent data corruption.
- Atomic promotion for review and write-log files ensures safe, race-free file updates.
- Introduces parent/child flock locking to prevent overlapping runs.
- Implements memory-as-data framing for improved prompt injection safety.
- Scripts now return exit code 2 for validation/atomic-write failures and 3 for lock contention.
- Adds atomic-write.sh helper library and updates documentation with new robustness guarantees.
v0.3.1
skillminer 0.3.1
- Clarified local vs external execution in documentation and metadata
- Added "git" to required binaries for improved consistency checks
- Updated environment variable documentation, now marking CLAWD_DIR as optional
- Improved warnings around FORGE_RUNNER=claude to highlight that it sends data externally
- Minor documentation cleanup and expanded environment usage notes
v0.3.0
skillminer 0.3.0 introduces richer scan summaries and manual trigger commands.
- Nightly scan summaries now include trend arrows, pending-age hints, and a live portfolio snapshot.
- New manual trigger commands: scan, write, full, status, and help—no need to remember full paths.
- Improved documentation for wrapper scripts and manual use cases.
- Added acceptance of both "letzen skillminer scan" and "letzten skillminer scan" as triggers.
v0.2.1
skillminer 0.2.1
- Adds automated cron announcement delivery after nightly scans and morning writes.
- Updates documentation in SKILL.md and user guides to reflect new notification flow.
- Minor text and formatting improvements across project docs.
- No changes to core skill triggers or command usage.
v0.2.0
**Major update with usability and documentation improvements.**
- Overhauled onboarding: clearer README, streamlined setup, and a new setup.sh installer script.
- Updated SKILL.md to clarify review flow, local-only operation, notifications, and runner/fallback options.
- Reduced and focused trigger phrases for command clarity.
- Prompts and scripts revised for better reliability and usability.
- Expanded and reworked documentation (README.md, USER_GUIDE.md) for easier first use.
v0.1.7
skillminer 0.1.7
- Renamed skill from "skill-miner" to "skillminer" for consistency.
- Updated triggers to correct German phrases (e.g., "was hat skill-miner gefunden" → "was hat skillminer gefunden").
- Updated documentation (README.md, USER_GUIDE.md, SKILL.md) to reflect new skill name and improved clarity.
- Incremented version to 0.1.7 in metadata and documentation.
- No functional behavior changes; this release focuses on naming and documentation consistency.
v0.1.6
skillminer 0.1.6
- Rewrote documentation for increased clarity and brevity, focusing on quick setup, natural language usage, and key differentiators.
- Updated README and SKILL.md with a concise overview, clearer installation steps, and streamlined instructions.
- No changes to core logic or schema; this release is documentation-focused to make onboarding and usage easier.
v0.1.5
- The skill now auto-detects its install location, reducing reliance on static paths and making custom installs easier.
- All references and scripts have been updated to use the auto-detected skill directory and correct naming conventions.
- Environment variable `CLAWD_DIR` is now used only for workspace memory and `_pending/` output, not for locating the skill code itself.
- Setup instructions, script paths, and file organization in all docs have been clarified and made more robust across deployment setups.
- No functional changes to review/promote logic; operational flow remains the same.
v0.1.4
skillminer 0.1.4
- Updated version to 0.1.4 in SKILL.md and related files.
- Documentation and metadata refreshed for clarity and consistency.
- No functional or trigger changes; skill behavior remains unchanged.
- Internal documentation (README.md, CHANGELOG.md) brought in sync with latest skill configuration.
元数据
常见问题
Skillminer 是什么？
Suggest reusable skills from recurring patterns in local memory files. Human review gate, drafts only to skills/_pending/, local-first runner with optional e... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 229 次。
如何安装 Skillminer？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install skillminer」即可一键安装，无需额外配置。
Skillminer 是免费的吗？
是的，Skillminer 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Skillminer 支持哪些平台？
Skillminer 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Skillminer？
由 Robby（@robbyczgw-cla）开发并维护，当前版本 v0.5.3。
推荐 Skills