thinking-partner
A deterministic thinking partner that challenges assumptions and applies mental models to sharpen decisions, solve problems, and think more clearly. Use this…
Refresh or query codebase intelligence (file graph, APIs, dependencies, architecture).
$ npx -y skills add SienkLogic/plan-build-run --skill intel --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/intelContext preview
The summary Claude sees to decide when to auto-load this skill.
Refresh or query codebase intelligence (file graph, APIs, dependencies, architecture).
name: intel description: "Refresh or query codebase intelligence (file graph, APIs, dependencies, architecture)." allowed-tools: Read, Write, Bash, Glob, Grep, Task, AskUserQuestion argument-hint: "[query <term>|refresh|status|diff]"
**STOP -- DO NOT READ THIS FILE. You are already reading it. This prompt was injected into your context by Claude Code's plugin system. Using the Read tool on this SKILL.md file wastes tokens. Begin executing Step 0 immediately.**
**Before ANY tool calls**, display this banner:
╔══════════════════════════════════════════════════════════════╗ ║ PLAN-BUILD-RUN ► INTEL UPDATE ║ ╚══════════════════════════════════════════════════════════════╝
Then proceed to Step 1.
Check if intel is enabled by reading `.planning/config.json` directly (do NOT use Bash for this -- `config get` throws on missing keys):
1. Read `.planning/config.json` 2. Check if `intel.enabled` is explicitly `false` 3. If `false`, display:
Intel system is disabled. Enable with: /pbr:config set intel.enabled true
Then **STOP**. Do not proceed further.
If enabled, not present, or no config file exists (all default to enabled), proceed to Step 2.
---
Parse the user's argument to determine the operation mode:
| Argument | Action | |----------|--------| | `query <term>` | Run inline query (Step 2a) | | `status` | Run inline status check (Step 2b) | | `diff` | Run inline diff check (Step 2c) | | `refresh` or no argument | Spawn intel-updater agent (Step 3) |
Run:
pbr-tools intel query <term>
Format and display the results. Show matching entries grouped by intel file. If no results found, suggest running `refresh` first.
**STOP** after displaying results. Do not spawn an agent.
Run:
pbr-tools intel status
Format and display staleness info for each intel file. Show last update times and whether files exist.
**STOP** after displaying status. Do not spawn an agent.
Run:
pbr-tools intel diff
Format and display changes since the last full refresh. Show which files have changed and summary of differences.
**STOP** after displaying diff. Do not spawn an agent.
---
**CRITICAL: Agent type rule** -- ALWAYS use `subagent_type: "pbr:intel-updater"`. NEVER use `general-purpose` or other non-PBR agent types. The PreToolUse hook will block non-PBR agents.
Spawn a Task with:
Wait for agent completion.
---
After the agent completes, run:
pbr-tools intel status
Display a summary showing:
---
1. DO NOT spawn an agent for query/status/diff operations -- these are inline CLI calls 2. DO NOT read agent .md files -- auto-loaded via subagent_type 3. DO NOT modify intel files directly -- the agent handles writes 4. DO NOT skip the config gate check 5. DO NOT use general-purpose agents -- always use `pbr:intel-updater`
Plan it. Build it. Run it. A Claude Code plugin for structured development with context-engineered agents.
Repo: SienkLogic/plan-build-run
A deterministic thinking partner that challenges assumptions and applies mental models to sharpen decisions, solve problems, and think more clearly. Use this…
Run audit, prioritize findings, auto-fix via quick tasks, test, and commit.
Review past Claude Code sessions for PBR workflow compliance and UX quality.
Run multiple phases hands-free. Chains discuss, plan, build, and verify automatically.
Manage backlog items — ideas not ready for active planning. Add, review, promote, or remove.
Start a new project. Deep questioning, research, requirements, and roadmap.