Skip to content
cccskills
FreePublished on GitHub

kb-init

Initialise a project for knowledge base use. Appends the [Knowledge Base] section to the project's CLAUDE.md, creates the library/ directory structure, optionally seeds with three example library files from the Agentic SDLC research, and reports next steps. Run once after installing sdlc-knowledge-base.

The English summary is being prepared. Showing the original description.

See how to install

Files included (7)

  • SKILL.md7.5 KB
  • templates/claude-md-section.md9.4 KB
  • templates/starter-pack/library/_shelf-index.md4.7 KB
  • templates/starter-pack/library/agent-suitability-rubric.md6.1 KB
  • templates/starter-pack/library/agentic-sdlc-options.md6.4 KB
  • templates/starter-pack/library/log.md845 B
  • templates/starter-pack/library/specification-formality-and-agent-performance.md7.6 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Initialise Knowledge Base

Set up a project to use the sdlc-knowledge-base plugin. This skill is idempotent — safe to run multiple times. It checks for existing setup before making changes.

Arguments

  • --with-starter-pack (default) — Seed library/ with three example library files synthesised from the Agentic SDLC research programme: agentic-sdlc-options.md, agent-suitability-rubric.md, specification-formality-and-agent-performance.md. New users get a queryable library on first install and can see the format end-to-end. The starter content is example only and meant to be replaced or extended.
  • --empty — Create the library directory structure but skip the starter pack. Use when you want a completely empty library to populate from scratch.
  • --library <path> — Initialise a library at this explicit path instead of ./library and skip the CLAUDE.md append (the section assumes one library at the default path). Use for isolated testing or a secondary library. See #209.

Steps

1. Verify the plugin is installed

Check that the sdlc-knowledge-base plugin is available. If not, report:

sdlc-knowledge-base plugin is not installed. Run:
  /plugin install sdlc-knowledge-base@ai-first-sdlc
Then re-run this skill.

2. Check for existing CLAUDE.md

The schema section gets appended to the project's CLAUDE.md. If CLAUDE.md does not exist, report:

CLAUDE.md not found in the project root. Create it first (even an empty file is fine), then re-run this skill.

3. Check for existing knowledge base section

Read CLAUDE.md and look for an existing ## Knowledge Base heading. If present, report:

[Knowledge Base] section already exists in CLAUDE.md. Skipping append.
If you want to reset the section, remove it manually and re-run this skill.

Then proceed to step 5 (directory creation) — it's still useful even if the section already exists.

4. Append the schema section template

Read the template from templates/claude-md-section.md (relative to this skill's directory; in plugin install this resolves to ${CLAUDE_PLUGIN_ROOT}/skills/kb-init/templates/claude-md-section.md).

Strip the HTML comment block at the top of the template (the <!-- ... --> block explaining what the template is for).

Append the remaining content to CLAUDE.md with one blank line of separation.

Report what was added:

Appended [Knowledge Base] section to CLAUDE.md (XX lines).

5. Create the library directory structure

Check whether library/ exists. If it does, skip the directory creation step (the starter pack copy in step 6 is still useful and safe).

If --library <path> is given, create the structure under <path> instead of ./library, and skip steps 2-4 (the CLAUDE.md append).

If it does not, create:

library/
library/raw/

Plus a placeholder library/.gitkeep so git tracks the empty directory.

Report:

Created library/, library/raw/

6. Optionally seed with the starter pack

Default mode (--with-starter-pack or no argument): copy the starter pack from the plugin install directory into the project's library/. The starter pack files live at ${CLAUDE_PLUGIN_ROOT}/skills/kb-init/templates/starter-pack/library/:

  • agentic-sdlc-options.md
  • agent-suitability-rubric.md
  • specification-formality-and-agent-performance.md
  • _shelf-index.md (with placeholder hashes)
  • log.md (with starter header and install entry)

For each file, check whether the destination already exists in library/. If yes, skip (don't overwrite). If no, copy.

Report:

Seeded library/ with starter pack:
  Copied: agentic-sdlc-options.md
  Copied: agent-suitability-rubric.md
  Copied: specification-formality-and-agent-performance.md
  Copied: _shelf-index.md (placeholder hashes — run kb-rebuild-indexes to regenerate)
  Copied: log.md

After copying, recommend the user run /sdlc-knowledge-base:kb-rebuild-indexes to replace the placeholder hashes in the shelf-index with real ones based on the actual file contents.

--empty mode: skip the copy entirely. Create only the directory structure plus an empty log.md with the starter header:

# Knowledge Base Log

Append-only chronological record of ingest, query, and lint events.
Format: `## [YYYY-MM-DD] <operation> | <subject>`

Report:

Created empty library structure (no starter pack).
Add raw sources to library/raw/ and run kb-ingest to populate.

7. Report next steps

Print a summary of what to do next.

For --with-starter-pack (default):

Knowledge base initialised with starter pack.

Next steps:
  1. Run /sdlc-knowledge-base:kb-rebuild-indexes to replace placeholder hashes with real ones
  2. Try a query: /sdlc-knowledge-base:kb-query "What does our research say about specification formality?"
  3. Review the example library files in library/ — replace or extend with content for your problem space
  4. Add your own raw sources to library/raw/ and run /sdlc-knowledge-base:kb-ingest <source> to integrate
  5. Read CLAUDE.md's new [Knowledge Base] section for the full conventions and workflow

For the format and design rationale, see:
  - The plugin README at plugins/sdlc-knowledge-base/README.md
  - The pattern document at docs/architecture/knowledge-base-pattern.md (in the framework repo)

For --empty:

Knowledge base initialised (empty).

Next steps:
  1. Add raw sources to library/raw/ (papers, articles, reports, conversation excerpts)
  2. Run /sdlc-knowledge-base:kb-ingest <source> to integrate them into structured library files
  3. After your first few library files exist, run /sdlc-knowledge-base:kb-rebuild-indexes to build the shelf-index
  4. Query the library with /sdlc-knowledge-base:kb-query "your question"
  5. Read CLAUDE.md's new [Knowledge Base] section for the full conventions and workflow

For the format and design rationale, see:
  - The plugin README at plugins/sdlc-knowledge-base/README.md
  - The pattern document at docs/architecture/knowledge-base-pattern.md (in the framework repo)

What this skill does NOT do

  • It does not write any library files. The user adds raw sources; kb-ingest produces library files from them.
  • It does not build the shelf-index. kb-rebuild-indexes does that, after library files exist.
  • It does not modify any agents or skills — they're shipped by the plugin install.
  • It does not configure environment validation integration. That's an explicit opt-in the user adds to CLAUDE.md after reviewing the schema section.

Idempotency

This skill is safe to run multiple times. On second invocation:

  • Existing [Knowledge Base] section in CLAUDE.md is detected and not duplicated
  • Existing library/ directory is detected and not recreated
  • The summary is still printed so the user knows where they are

Error handling

  • CLAUDE.md missing → preflight failure with guidance (step 2)
  • Plugin not installed → preflight failure with install command (step 1)
  • Template file unreadable → fail with explicit path so the user can check the install
  • Permission denied creating library/ → report the error and stop; do not continue with partial setup

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Generate a GitHub Actions workflow for Android — JDK + Gradle caching, build, unit + instrumented tests, Android Lint, run the Android pre-flight checks, and (optionally) sign + upload a signed app bundle to a Play track. Use when setting up CI for an Android app.

The English summary is being prepared. Showing the original description.

SteveGJones/ai-first-sdlc-practices41Updated Oct 5, 2026

Validated Google Play release — runs the pre-flight checks and clears the Play policy gates, builds a signed app bundle with a git-stamped versionCode, uploads to a track, and walks staged rollout (with halt/roll-forward). Use when releasing an Android app to Play.

The English summary is being prepared. Showing the original description.

SteveGJones/ai-first-sdlc-practices41Updated Oct 5, 2026

Scaffold a new Android app with SDLC and Play-safe defaults — Kotlin/Compose Gradle project with version catalog, convention plugins, module layout, a targetSdk that meets the Play mandate, R8 on release, and git-ignored signing so it passes the pre-flight checker from day one. Use when starting a new Android project.

The English summary is being prepared. Showing the original description.

SteveGJones/ai-first-sdlc-practices41Updated Oct 5, 2026

Set up Android app signing safely — generate an upload keystore, enroll in Play App Signing, and wire a Gradle release signingConfig that reads secrets from a git-ignored keystore.properties / CI (never committed). Use when configuring signing for a Play release.

The English summary is being prepared. Showing the original description.

SteveGJones/ai-first-sdlc-practices41Updated Oct 5, 2026

Pick an existing workflow for a task or author a new one — recommends workflow + team formation, or generates a fresh workflow YAML with command prompts and team validation.

The English summary is being prepared. Showing the original description.

SteveGJones/ai-first-sdlc-practices41Updated Oct 5, 2026

Guide the team through writing a change-impact record for IEC 62304 / FDA / ISO 26262 ASIL C+ projects. Surfaces the affected REQ/DES/TEST/CODE artefacts and captures human-attested verification approach.

The English summary is being prepared. Showing the original description.

SteveGJones/ai-first-sdlc-practices41Updated Oct 5, 2026

See all skills by SteveGJones

Report a problem with this skill