SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillslcs-research

lcs-research

Writing & research

View repository

Updated 10-16 00:22

About this skill

LCS Research Skill

Shared Coding Contract: Refer to ../lcs-shared/contract.md for folder conventions, Handoff format, and token optimization.

Purpose: Investigate a question against high-trust primary sources. Capture findings in a cited Markdown file.

Chain of Truth Level: Standard

OKF Frontmatter & Writing Safety

When creating artifacts, include YAML frontmatter following the schema in ../lcs-shared/contract.md. Follow Artifact Writing Safety rules: generate content first, write one file, verify, stop on failure.

Trigger

Activate when user wants to: research a library/API, look up documentation, verify how something works, or gather evidence for a technical decision.

Behavior Checklist

  1. Identify the research question and required primary sources (official docs, source code, RFCs).

  1. Spin up a background investigation process (or sequential deep-search if background agents unavailable).

  2. Rule: Follow every claim back to its primary source. Do NOT cite secondary blog posts or summaries unless the primary source is unavailable. If primary source is unavailable, note it explicitly.

  3. Write findings to .lcs/work-items/{ts}-{slug}/research/<topic>.md.

  4. Include a ## Citations section with exact URLs, file paths, and line numbers.

  5. End with Handoff passing the research artifact back to the invoking skill (e.g., lcs-explore, lcs-wayfinder, lcs-toprd).

Output

  • .lcs/work-items/{ts}-{slug}/research/<topic>.md with structured findings and citations

Chain of Truth Report

Level

Standard

Sources Checked

  • <Primary sources consulted, with URLs/paths>
  • .lcs/state.md (if present)

Assumptions

  • <label each [verified] or [unverified]>

Actions Taken

Verification

  • <Findings written; citations verified against primary sources>

Report

<1-3 sentence summary with confidence rating>

Handoff

Next recommended skill: {invoking-skill} Next file to read: .lcs/work-items/{ts}-{slug}/research/.md Current phase: research Current confidence: <low/medium/high> Blocking questions: Risks to carry forward: Source of Truth Bundle: .lcs/state.md, research/.md Must Preserve IDs: None Unresolved IDs: None Suggested next command: Kembali ke skill pemanggil dan lanjut flow

Chain of Truth Level

Level: Standard

This skill follows the LCS Chain of Truth protocol at the declared level.