Skip to content
mcp/skillhub

Agent Skill

Kiro Discovery skill

kiro-discoverygotalab/cc-sddtools/cc-sdd/templates/agents/claude-code-skills/skills/kiro-discovery/SKILL.md

Entry point for new work. Determines the best action path or work decomposition (update existing spec, create new spec, mixed decomposition, or no spec needed) and refines ideas through structured dialogue.

3.7kMIT

always on

44 tokname + description

once used

2,928 tokSKILL.md body

bundled files

0

repo stars

3.7kGitHub

Install the Kiro Discovery skill

Skills CLI (Claude Code, Codex, Cursor and others)

npx skills add https://github.com/gotalab/cc-sdd --skill kiro-discovery

Claude Code, by hand (all projects)

git clone --depth 1 https://github.com/gotalab/cc-sdd.git /tmp/cc-sdd
mkdir -p ~/.claude/skills
cp -R /tmp/cc-sdd/tools/cc-sdd/templates/agents/claude-code-skills/skills/kiro-discovery ~/.claude/skills/kiro-discovery

For one project only, copy the folder into .claude/skills/ in that project instead. It is marked disable-model-invocation, so the agent won't pick it on its own: you run it as /kiro-discovery.

What it costs your context

Agent Skills load in three steps. Tokens counted with the o200k tokenizer; your model's count will differ a little.

WhenWhat loadsCost
Every sessionName and description44 tok
When the agent uses itThe SKILL.md instructions2,928 tok
Only if openedNo bundled files—

Until it is used, kiro-discovery costs about 31× less than the median MCP server we measured (1,363 tokens on every turn). Installing all 150 skills from gotalab/cc-sdd would load 3,459 tokens of descriptions. Skills vs MCP study

Checks

Our checks found nothing to flag: it follows the Agent Skills specification's naming and size guidelines and doesn't pipe downloads into a shell.

Pre-approved tools (allowed-tools): Read, Write, Glob, Grep, Agent, WebSearch, WebFetch, AskUserQuestion. The agent can use these without asking while the skill is active.

Checked 2026-10-22 · method

What's inside

The headings of its SKILL.md. Read the full file on GitHub.

  • kiro-discovery Skill
  • Core Mission
  • Execution Steps
  • Step 1: Lightweight Scan
  • Step 2: Determine Action Path
  • Step 3: Deep Context Loading
  • Step 4: Understand the Idea
  • Step 5: Propose Approaches
  • Step 6: Refine and Confirm
  • Step 7: Write Files to Disk
  • Step 8: Suggest Next Steps
  • Critical Constraints
  • Safety & Fallback

More skills from gotalab/cc-sdd

All skills in this repo
  1. cc-sdd-new-agentgotalab/cc-sdd

    Add or extend coding-agent support in cc-sdd by executing the SOP in docs/cc-sdd/sop-new-agent.md end-to-end. Use when introducing a new agent, adding a subagent-capable variant, or evaluating migration of an existing supported agent to skills-based templates.

    3.7k63 tok idle · 982 in use
  2. kiro-debuggotalab/cc-sdd

    Investigate implementation failures using root-cause-first debugging. Use when an implementer is blocked, verification fails, or repeated remediation does not converge.

    3.7k33 tok idle · 1,155 in use
  3. kiro-debuggotalab/cc-sdd

    Investigate implementation failures using root-cause-first debugging. Use when an implementer is blocked, verification fails, or repeated remediation does not converge.

    3.7k33 tok idle · 1,143 in use
  4. kiro-discoverygotalab/cc-sdd

    Entry point for new work. Determines the best action path or work decomposition (update existing spec, create new spec, mixed decomposition, or no spec needed) and refines ideas through structured dialogue.

    3.7k44 tok idle · 2,839 in use
  5. kiro-discoverygotalab/cc-sdd

    Entry point for new work. Determines the best action path or work decomposition (update existing spec, create new spec, mixed decomposition, or no spec needed) and refines ideas through structured dialogue.

    3.7k44 tok idle · 2,839 in use
  6. kiro-discoverygotalab/cc-sdd

    Entry point for new work. Determines the best action path or work decomposition (update existing spec, create new spec, mixed decomposition, or no spec needed) and refines ideas through structured dialogue.

    3.7k44 tok idle · 2,851 in use
  7. kiro-implgotalab/cc-sdd

    Implement approved tasks using TDD with subagent dispatch. Runs all pending tasks autonomously or selected tasks manually.

    3.7k27 tok idle · 4,101 in use
  8. kiro-implgotalab/cc-sdd

    Implement approved tasks using TDD with native subagent dispatch. Runs all pending tasks autonomously or selected tasks manually.

    3.7k28 tok idle · 4,053 in use