English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

COLLEAGUE.SKILL: Distilling Human Expertise into Inspectable, Installable Agent Skills

Forum topic · 小凯 · 2026-06-03

Summary

COLLEAGUE.SKILL is a framework from Shanghai AI Lab that converts raw human traces—chat logs, documents, emails, interviews—into standardized Agent Skill packages. Its core contribution is a dual-track representation that separates work methods and mental models (capability track, work.md) from communication style and interaction rules (behavior track, persona.md), enabling independent or combined invocation. The system ships three governance-aware presets: colleague, celebrity, and relationship, each with tailored evidence scope, consent requirements, and control mechanisms. Generated skills follow the SKILL.md / meta.json artifact standard, are portable across Claude Code, OpenClaw, Codex, and Hermes, and support natural-language correction with versioned rollback. A quality-checker audits mind-model coverage, limitation labels, expression patterns, internal contradictions, source URLs, and copyright signals. The project has spawned an open ecosystem of 215+ public skills, 165 contributors, and 100k+ gallery stars. Rather than cloning identity, COLLEAGUE.SKILL packages reusable expert judgment as inspectable, correctable, governable artifacts.

Key points

  • Problem framing. LLM agents are shifting from one-off task execution to carrying reusable work contexts—code-review standards, expert heuristics, mental models, interaction patterns—but this knowledge lives scattered across chats, design docs, emails, and meeting notes. Memory systems capture fragments without structure; persona systems conflate knowledge, judgment, and tone; skill frameworks lack an end-to-end workflow from raw traces to skills.
  • Core insight. Convert selected human traces into inspectable, correctable, installable, governable skill packages—not hidden memory stores or identity-replication claims.
  • Dual-track representation. Skills split into a Capability Track (work methods, mental models, decision heuristics, technical standards in work.md) and a Behavior Track (communication style, interaction rules, boundary constraints, correction logs in persona.md). The split enables three invocation modes: /{character}-{slug}-work, /{character}-{slug}-persona, and /{character}-{slug} (combined).
  • Three application presets. The same underlying workflow is parameterized by scenario:
  • *colleague*: Slack, Feishu, DingTalk, WeChat, email, docs; focuses on work knowledge and enterprise permissions.
  • *celebrity*: interviews, talks, writings, subtitles, decision logs; focuses on source boundaries, fact-checking, copyright.
  • *relationship*: private chats, diaries, letters; focuses on informed consent, local control, deletion paths.
  • Artifact structure. A standard skill package contains SKILL.md, work.md, persona.md, work_skill.md, persona_skill.md, manifest.json, and meta.json (lifecycle state: version, correction count, sources). Key properties: Portable (Claude Code, OpenClaw, Codex, Hermes), Inspectable, Composable, Correctable (natural-language patches, versioned rollback), Governable.
  • Creation workflow. User supplies an alias and source materials (Feishu API, DingTalk browser mode, Slack API, WeChat SQLite export, .eml/.mbox, PDF, images, Markdown) → collector/parser → analyzer → builder → writer producing the packaged artifact.
  • Correction workflow. Natural-language feedback (e.g., "he would not say that") is classified as capability or persona, packaged as a Markdown patch or normalized {scene, wrong, correct} record, versioned with rollback (list, rollback, backup, clean).
  • Celebrity research toolchain. Six-dimensional research framework: Works, Interviews, Decisions, Expression DNA, External Evaluations, Timeline. Scripts: download_subtitles.sh, srt_to_transcript.py, merge_research.py, quality_check.py. The checker audits mind-model coverage, limitation labeling, expression patterns, internal contradictions, source URLs, and copyright safety signals.
  • Ecosystem (as of 2026-05-28). 215 public skills, 55 meta-skills, 165 contributors, 100k+ cumulative gallery stars.
  • Representative cases. Andrej Karpathy-style skill emphasizing scaffolding and strict evaluation over raw demos; ByteDance backend-engineer skill enforcing N+1 fixes and standard response formats; relationship skill with conversational pushback.
  • Position vs. related work. Distinct from Voyager/SkillX/SkillGen/AutoSkill (skills from execution traces), RAG/LaMP/PersonaAgent (retrieval-based memory), Character-LLM/RoleLLM/SOTOPIA (behavioral cloning), and Agent Skills Spec/Claude Skills (format only). COLLEAGUE.SKILL is the first to provide an end-to-end trace-to-skill workflow with separated capability/persona tracks.
  • Stated limitations. Does not pursue behavioral fidelity; does not solve identity substitution; output quality depends on source quality; user corrections may make contested traces appear more certain.
  • Responsible deployment requirements. Informed participation, scoped collection, access control, retention limits, non-coerced use, opt-in gallery with submitter verification, moderation, takedown, and source-boundary labels for celebrity/relationship presets.
  • Resources

  • Paper: https://arxiv.org/abs/2605.31264
  • Code: https://github.com/titanwings/colleague-skill
  • Project page: https://titanwings.github.io/colleague-skill-site/

Tags

#ai-agents#knowledge-distillation#skill-framework#persona-modeling#expert-systems#open-source#shanghai-ai-lab#agent-governance

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177980765