- Skills
- 42
- 仓库
- 1
- 职业覆盖
- 0
- 场景覆盖
- 0
- Stars
- ★ 1
HexingBot 的 Skills 42
unknown-skill
A demo skill with no license signal at all, used to exercise the "unknown" classification branch of the human license gate (TASK-135).
spoofed-license-skill
A demo skill used to test TASK-150's SPDX-header-vs-LICENSE-file conflict surfacing.
permissive-skill
A demo skill used to test assimilation of a permissively-licensed third party skill.
malicious-skill
A fixture skill deliberately carrying risky content patterns, used only to exercise src/skill-scan.js (TASK-122). Never assimilated for real.
copyleft-skill
A demo skill used to test the human license gate for a copyleft third party skill.
watch
Watch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video. (assimilated for Hivemind)
tech-training-template
Template for the Researcher to use when training the team on a new tech stack. Copy this folder to .claude/skills/<stack-name>/ and replace the placeholders. The frontmatter description must clearly state WHEN to load this skill so Claude can decide relevance via progressive disclosure.
orchestrator-routing
Always load at the start of every orchestrator chat in an hivemind project. Carries the non-negotiable RESUME-FIRST session-resume contract, the first-chat init routing rule, and the full operational manual for the Orchestrator role (delegation protocol, developer spawn modes, reviewer isolation, UAT procedure, ticket-update protocol, tier rubric).
mcp-server
How to build, bundle, register, and test a minimal stdio MCP (Model Context Protocol) server in Node.js with the official @modelcontextprotocol/sdk. Load this when implementing or modifying an MCP server (src/mcp-server.js, .mcp.json), when wrapping internal functions as MCP tools, when bundling the SDK with esbuild into a CJS artifact, or when the eventual Atlassian-MCP migration swaps the task-store backend. Triggers on @modelcontextprotocol/sdk, McpServer, StdioServerTransport, registerTool, InMemoryTransport, or ${CLAUDE_PROJECT_DIR} in an MCP context.
manifest-verifier
Run the objective cross-manifest coverage checks (npm run check:manifests, backed by src/manifest-verify.js) and emit a pass/fail matrix to reviews/VERIFY.md. Optional consistency check for a developer who chose to write the impl-* manifests — not a gate before code (TASK-230, 2026-09-16 human decision). Load whenever that consistency check is wanted.
impl-state-schemas
Generate the STATE_SCHEMAS manifest — server/query state (query-key factory) + UI/local state stores + cache-invalidation rules. Optional tool a developer may use anytime when implementing state management; not required before code (TASK-230, 2026-09-16 human decision). Vendored Spine skill; see src/manifest-policy.js.
impl-screen-specs
Generate the SCREEN_SPECS manifest — a per-screen specification (components, data, API calls, user actions, states, i18n). Optional tool a developer may use anytime to blueprint screens; not required before code (TASK-230, 2026-09-16 human decision). Vendored Spine skill; see src/manifest-policy.js.
impl-project-structure
Generate the PROJECT_STRUCTURE manifest — annotated source tree, module boundaries, import conventions, and ADR-decided stack assumptions (language/framework-agnostic). Optional tool a developer may use anytime when establishing or extending the project skeleton; not required before code (TASK-230, 2026-09-16 human decision). Vendored Spine skill; see src/manifest-policy.js.
impl-component-catalog
Generate the COMPONENT_CATALOG manifest — shared UI components, one canonical entry per input type (date, select, text, …); props, behavior, i18n. Enforces zero-duplicate UI consistency. Optional tool a developer may use anytime when building shared components; not required before code (TASK-230, 2026-09-16 human decision). Vendored Spine skill; see src/manifest-policy.js.
impl-block-tasks
Generate the BLOCK_TASKS manifest — every work block broken into concrete developer tasks with testable + observable acceptance criteria, dependencies, and blocking gaps. Optional tool a developer may use anytime when decomposing an epic/block into tickets; not required before code (TASK-230, 2026-09-16 human decision). Vendored Spine skill; see src/manifest-policy.js.
impl-api-contracts
Generate the API_CONTRACTS manifest — an endpoint table by module (method, path, auth, request/response shapes, cache keys, invalidation). Optional tool a developer may use anytime when implementing endpoints or data fetching; not required before code (TASK-230, 2026-09-16 human decision). Vendored Spine skill; see src/manifest-policy.js.
hive-self-improve-current-project
Load when improving the QUALITY of an already-shipped component of THIS project (your own src/, tests/, tasks/ — not the hivemind framework) — simplification, duplication and dead-code removal, test-coverage gaps, doc/consistency drift, developer ergonomics, and performance — by running the real code to ground every finding, then converting each into a right-tiered ticket. NOT for security or trust-boundary hardening (that is hive-adversarial-improve-current-project) and NOT for routine single-diff review (that is the reviewer agent). Triggers on "hive-self-improve-current-project this", "improve THIS project", "find quality gaps in <component>", "what can we simplify/clean up in <component>".
hive-adversarial-improve-current-project
Load when hardening a specific component/pipeline of THIS project (your own src/, tests/, tasks/ — not the hivemind framework) by stress-testing it against difficult and hostile inputs — especially a trust boundary where untrusted content crosses into trusted execution (auth/validation seams, installers, parsers of fetched or user-supplied content). Triggers on "hive-adversarial-improve-current-project this", "harden this component", "stress-test this boundary", "find failure modes in <component>".