← Browse

@longngn204/update-local-coding-agent

B

Safely update an existing Local Coding Agent clone while preserving customer config, tunnel-client, and secrets.

skillclaude

Install

agr install @longngn204/update-local-coding-agent --target claude

Writes 1 file into .claude/skills/, pinned to git-ba51f320.

  • .claude/skills/update-local-coding-agent/SKILL.md

Document


name: update-local-coding-agent description: Safely update an existing Local Coding Agent clone while preserving customer config, tunnel-client, and secrets.

Update Local Coding Agent

Use this when a customer wants their local clone updated to the latest GitHub version.

Rules

  • Do not delete the customer workspace.
  • Do not delete tools/tunnel-client*.
  • Do not print, commit, or upload API keys, tunnel IDs, tokens, or local config.
  • Do not run git reset --hard, git clean, or destructive commands unless the customer explicitly approves.
  • If there are local changes, summarize them and ask before continuing.

Steps

  1. Find the local local-coding-agent folder.
  2. Run git status --short --branch.
  3. If local changes exist, stop and ask before updating.
  4. Fetch:
    • git fetch origin main --tags
  5. Show incoming changes:
    • git log --oneline --decorate --max-count=10 HEAD..origin/main
  6. Update safely:
    • git pull --ff-only origin main
  7. Reinstall dependencies:
    • Windows: scripts\lca.cmd install
    • macOS/Linux: bash scripts/lca install
  8. Validate:
    • node --check scripts/local-coding-agent.mjs
    • node --check scripts/network-doctor.mjs
    • node scripts/validate-skills.mjs
  9. Run doctor/status:
    • Windows: scripts\lca.cmd doctor and scripts\lca.cmd status
    • macOS/Linux: bash scripts/lca doctor and bash scripts/lca status
  10. Restart only if the customer wants the agent running:
  • stop, then start with the CLI wrapper for the OS.

Report Back

Return:

  • current commit
  • current version from /healthz if running
  • MCP URL
  • dashboard URL
  • workspace path
  • mode and policy
  • tunnel status
  • any failed check and exact next command

Trustgrade B

  • passBody integrity

    Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.

  • passType matchnot applicable to this artifact type

    Whether the artifact is really the kind of thing its metadata claims it is.

  • passFreshness

    How long since the source repository was last pushed to.

  • passPrompt injection

    Scans the artifact's own text for instructions aimed at your agent rather than at you.

  • warnLicensecopyleft/unknown — index-and-link only

    Whether the source repository declares an SPDX license permissive enough to redistribute.

How the grade is calculated

Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:

  • Aevery check passed
  • Bone warning
  • Ctwo warnings
  • Dprompt injection or body integrity failed, or three warnings
  • Fone of those failed, and something else is wrong

These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.

Versions

  • git-ba51f320e41d2026-07-31